Abstract
As cells progress through carcinogenesis, the associated exponential expansion of genetic and molecular aberrations and resultant heterogeneity make therapeutic success increasingly unattainable. Therapeutic intervention at early stages of carcinogenesis that occurs within the primary organ and in the face of a lower burden of molecular aberrations, constitutes a basic tenet of cancer chemoprevention, and provides a situation that favors a greater degree of therapeutic efficacy compared with that of advanced cancer. A longstanding barrier to chemoprevention relates to the requirement for essentially no systemic toxicity, and the fact that when large numbers of people are treated, the emergence of systemic toxicity is almost universal. A rational means to address this in fact relates to a second basic tenet of the chemopreventive strategy: the focus of therapeutic intervention is to disrupt a process that is in essence localized to a single organ. Based upon this consideration, a strategy which is based upon local delivery of therapeutics to an at-risk organ will achieve therapeutic efficacy while avoiding systemic delivery and its associated toxicity. This article will review the rationale for undertaking such an approach, describe successful clinical achievements based on this strategy, describe ongoing efforts to expand the impact of this approach, and together will highlight the high impact that this approach has already had on the field as well as its extremely high potential for future impact. Cancer Prev Res; 10(1); 14–35. ©2016 AACR.
Introduction
The prevention of cancer through therapeutic means is based upon two fundamental principles. The first relates to a central tenet of cancer biology. As normal cells progress through the process of carcinogenesis, the number and variety of genetic, molecular, and resultant functional aberrations increases in an essentially exponential fashion. As a direct consequence of this, there is a similar increase in the number of dysregulated processes that need to be therapeutically reverted to affect clinical control of the disease. The second fundamental principle relates to a basic aspect of pharmacology. Any given therapeutic agent has a limited spectrum of pharmacologic targets that it acts upon. Prototypically, a given agent exerts its desired therapeutic effects by modulating only a single target, and in practice, this is typically the case. A combined consideration of these two principles underlies the logic that therapeutic efficacy increases as it is applied to successively earlier stages in carcinogenesis. When applied to very early stages of carcinogenesis, the goal of therapeutic intervention is cancer prevention.
These fundamental principles are best exemplified by considering breast cancer. In the setting of metastatic cancer, even highly toxic combination chemotherapy is unable to affect long-term control and cure. In localized disease, therapy can affect cure, but requires utilization of intensive combination chemo- and hormonal therapy, coupled with local surgical and/or radiotherapy. In this situation, there is a relatively low burden of systemic disease present. In fact, it is so low that it cannot be clinically detected. Even in this situation, systemic combination therapy only prevents the future emergence of disease in a subset of patients. However, in asymptomatic women who do not have clinical evidence of breast cancer, but are at risk for such, utilization of a single agent, tamoxifen, will decrease the probability of developing breast cancer by 50% (1). Therefore, by moving the therapeutic target down the spectrum of carcinogenesis, that is, to individuals at risk for disease, high levels of disease control can be achieved. This relates to at-risk tissue harboring a lower burden of dysregulated processes, compared with metastatic disease that requires pharmacologic manipulation.
Founded in sound scientific principles, it is the pursuit of this high potential for chemoprevention which serves to drive the field. It is the dream of researchers, physicians, and patients alike, to have the ability to “take a pill”, and to thereby prevent cancer. Despite aggressive pursuit of these goals, our ability to achieve them has fallen disappointingly short. Across all subtypes of cancer, the FDA has approved only 12 agents for cancer risk reduction (Table 1). In addition to these agents, several interventions target viruses and bacteria known to be associated with cancer development (Table 2).
US FDA approved agents for cancer risk reduction
Drug . | Cancer type . | Mechanism . | Targeted cohort . |
---|---|---|---|
Tamoxifen | Breast | SERM: Selective estrogen receptor modulator | Women with DCIS (ductal carcinoma in situ) after breast surgery and radiation |
Women at high risk for breast cancera | |||
Raloxifene | Breast | SERM | Postmenopausal women at high risk for invasiveb breast cancer |
Porfimer sodium + photodynamic therapy (PDT) & omeprazole | Esophageal | Production of oxygen-free radicals kills nearby cancer cells of precancerous tissues after exposure to light | Males and females with high-grade dysplasia in Barrett esophagus |
Aspirin | Colorectal | Induces apoptosis and interrupts prostaglandin production through inhibition of cyclooxygenase enzymes | Adults ages 50–59 yearsc |
Celecoxib | Colorectal | Induces apoptosis and interrupts prostaglandin production through selective inhibition of the cyclooxygenase-2 enzyme | Males and females ≥18 years with FAP (Familial adenomatous polyposis) |
Bacillus Calmette-Guerin (BCG) | Bladder | Mechanism unknown; induces local immune reaction against tumor | Males and females with CIS (carcinoma in situ) of the urinary bladder |
Valrubicin | Bladder | Interferes with DNA synthesis and leads to cell death | Males and females with BCG-refractory CIS |
Fluorouracil | Skin | Interferes with DNA synthesis and leads to cell death | Males and females with multiple actinic keratoses |
Diclofenac sodium 3% | Skin | Mechanism unknown | Males and females with actinic keratoses |
5-aminolevulinic acid + PDT | Skin | Solution kills precancerous cell when exposed to light | Males and females with actinic keratoses of the face or scalp |
Imiquimod | Skin | Enhances immune response and promotes apoptosis | Immunocompetent males and females with actinic keratoses of the face or scalp |
Ingenol mebutate | Skin | Induces primary necrosis | Males and females with actinic keratoses on the face, scalp, trunk and extremities |
Drug . | Cancer type . | Mechanism . | Targeted cohort . |
---|---|---|---|
Tamoxifen | Breast | SERM: Selective estrogen receptor modulator | Women with DCIS (ductal carcinoma in situ) after breast surgery and radiation |
Women at high risk for breast cancera | |||
Raloxifene | Breast | SERM | Postmenopausal women at high risk for invasiveb breast cancer |
Porfimer sodium + photodynamic therapy (PDT) & omeprazole | Esophageal | Production of oxygen-free radicals kills nearby cancer cells of precancerous tissues after exposure to light | Males and females with high-grade dysplasia in Barrett esophagus |
Aspirin | Colorectal | Induces apoptosis and interrupts prostaglandin production through inhibition of cyclooxygenase enzymes | Adults ages 50–59 yearsc |
Celecoxib | Colorectal | Induces apoptosis and interrupts prostaglandin production through selective inhibition of the cyclooxygenase-2 enzyme | Males and females ≥18 years with FAP (Familial adenomatous polyposis) |
Bacillus Calmette-Guerin (BCG) | Bladder | Mechanism unknown; induces local immune reaction against tumor | Males and females with CIS (carcinoma in situ) of the urinary bladder |
Valrubicin | Bladder | Interferes with DNA synthesis and leads to cell death | Males and females with BCG-refractory CIS |
Fluorouracil | Skin | Interferes with DNA synthesis and leads to cell death | Males and females with multiple actinic keratoses |
Diclofenac sodium 3% | Skin | Mechanism unknown | Males and females with actinic keratoses |
5-aminolevulinic acid + PDT | Skin | Solution kills precancerous cell when exposed to light | Males and females with actinic keratoses of the face or scalp |
Imiquimod | Skin | Enhances immune response and promotes apoptosis | Immunocompetent males and females with actinic keratoses of the face or scalp |
Ingenol mebutate | Skin | Induces primary necrosis | Males and females with actinic keratoses on the face, scalp, trunk and extremities |
a"High risk" defined as women at least age 35 years with a 5-year predicted risk of breast cancer ≥ 1.67% as calculated by the Gail model.
b“High risk” defined as at least one breast biopsy showing lobular carcinoma in situ (CIS) or atypical hyperplasia, one or more first-degree relatives with breast cancer, or a 5-year predicted risk of breast cancer ≥ 1.66% (based on the modified Gail model).
cLow-dose aspirin use for the primary prevention of cardiovascular disease and colorectal cancer in adults ages 50–59 years who have a 10% or greater 10-year CVD risk, are not at increased risk for bleeding, have a life expectancy of at least 10 years, and are willing to take low-dose aspirin daily for at least 10 years.
Vaccines and antimicrobial interventions likely beneficial in cancer prevention
Infectious organism . | Available intervention (s) . | Cancer type . |
---|---|---|
Human papillomavirus (HPV) | Gardasil 9 vaccine | Cervix |
- HPV - 6, 11, 16, 18, 31, 33, 45, 52, 58 | Vulva | |
Cervarix vaccine | Vagina | |
- HPV - 16, 18 | Anusa | |
Hepatitis B virus | Hepatitis B vaccine | Hepatocellular carcinoma |
Interferon therapy | ||
Nucleoside analogues | ||
Hepatitis C virus | Nucleoside analogues | Hepatocellular carcinoma |
Interferon therapy | ||
Human immunodeficiency virus (HIV) | Antiretroviral therapy | Non-Hodgkin lymphoma |
- Human herpes virus 8 (HHV-8) | Kaposi Sarcoma | |
Helicobacter pylori | Antibiotics | Stomach |
Schistomiasis | Antischistosomal | Bladder |
Clonorchis sinensis, Opisthorchis viverrini (liver flukes) | Antihelmenthic | Cholangiocarcinoma |
Infectious organism . | Available intervention (s) . | Cancer type . |
---|---|---|
Human papillomavirus (HPV) | Gardasil 9 vaccine | Cervix |
- HPV - 6, 11, 16, 18, 31, 33, 45, 52, 58 | Vulva | |
Cervarix vaccine | Vagina | |
- HPV - 16, 18 | Anusa | |
Hepatitis B virus | Hepatitis B vaccine | Hepatocellular carcinoma |
Interferon therapy | ||
Nucleoside analogues | ||
Hepatitis C virus | Nucleoside analogues | Hepatocellular carcinoma |
Interferon therapy | ||
Human immunodeficiency virus (HIV) | Antiretroviral therapy | Non-Hodgkin lymphoma |
- Human herpes virus 8 (HHV-8) | Kaposi Sarcoma | |
Helicobacter pylori | Antibiotics | Stomach |
Schistomiasis | Antischistosomal | Bladder |
Clonorchis sinensis, Opisthorchis viverrini (liver flukes) | Antihelmenthic | Cholangiocarcinoma |
aOnly Gardasil 9 is approved in males ages 9 through 15 years.
While many factors go into considering the clinical value of any agent, two of them predominate: efficacy and resultant clinical benefit, and toxicity and resultant clinical risk. With a chemoprevention strategy, long-term treatment of large numbers of people is undertaken so as to prevent the development of adverse consequences in a relatively small subset of treated individuals at a relatively distant time in the future. The adverse consequences we are seeking to pharmacologically prevent, for example, emergence of cancer, can have a high degree of negative impact upon an individual. Despite this very impactful goal at the level of the individual who does not have to face the advent of cancer therapy, the fact is that at the population level, the theoretical optimal clinical benefit that can be achieved is not high.
Because of this, there is the expectation that an acceptable chemopreventive agent will have little-to-no toxicity. This attribute has been highly problematic for the field, and serves as a longstanding prohibitory barrier. This relates to the fundamental fact that any substance taken into the body has the ability to induce toxicity, even if it is a highly effective therapeutic with a clear and substantial risk:benefit ratio. This is framed in the famous saying by Paracelsus, “Poison is in everything, and no thing is without poison. The dosage makes it either a poison or a remedy.” The desire for a chemopreventive agent to have no toxicity is therefore diametrically opposed to the fact that any substance taken into the body can induce toxicity. In practice, the barrier to achieving minimal toxicity with chemopreventive agents is heightened by the requirement that they are administered on a continuous basis over an extended period of time.
The avoidance of systemic toxicity is therefore a central goal of cancer chemoprevention, and constitutes a longstanding barrier to the field. The expectations are very high. Consider the case of tamoxifen. Despite exerting a dramatic effect upon a very common cancer in a high-risk cohort that can be readily identified, its widespread use has been limited by its side effects, which include hot flashes, thromboembolic events, and increased risk of uterine cancer (2). Even though the latter was of low incidence and readily detected at a highly curable stage, it was considered unacceptable to most patients and of concern to most practitioners. Similar situations exist for other risk reduction agents. Examples include use of sunscreen for prevention of skin cancer, where cosmetic-related concerns have limited use, and use of non-steroidal anti-inflammatory drugs (NSAID) for the prevention several cancers (3, 4), where increased risk for gastrointestinal bleeding and hemorrhagic stroke has limited the drug class from gaining approval as a population-health–based intervention.
Rapid advances in our understanding of carcinogenesis are enhancing our ability to deploy targeted therapy with increasingly impressive efficacy. The same advances, when coupled with rapidly emerging technologies, are together allowing us to redefine and risk-stratify individuals who may benefit from therapy with increased accuracy. The former will serve to increase agent specificity, while the latter will allow us to spare those at low risk from the adverse effects of toxicity when their overall benefit from such is small. Both will serve to improve the risk:benefit ratio, just as it has done for advanced cancer. Even with these advances it will not be possible to avoid toxicity. Many of our targeted therapeutics, inclusive of those considered highly selective, induce toxicity to non-target tissues. Despite therapies being designated as targeted, this situation is not surprising, and relates to the above concept that all substances having the potential to induce toxicity. Furthermore, the ability of targeted agents to pharmacologically alter a singular specific target protein varies widely across the spectrum of so called targeted agents. Finally, the protein being targeted may also be expressed in normal tissue, whereupon its pharmacologic modulation induces toxicity in the organ in which that tissue resides.
Reasonable strategies that have been proposed to reduce systemic toxicity advocate combining different therapies, alternate dosing schedules, and lower doses of cancer chemopreventive agents. Combining various effective chemopreventive agents at lower doses than when used individually has been proposed as an approach that could synergistically target multiple oncogenic pathways and better address tumor heterogeneity and selection pressure, while decreasing the risk for toxicity. Intermittent, instead of continuous chemoprevention, could also reduce adverse effects of the same chemopreventive agents used continuously in systemic therapy while retaining efficacy. Alternately, a low-dose chemoprevention strategy is founded on the belief that prevention of cancer cells from developing should require a lower dose than eradicating established tumor cells (5). These approaches offer plausible ways to minimize toxicity while preserving the potency of a given chemopreventive agent and set the stage for more definitive clinical trials that could potentially distinguish the overall value between these alternate strategies across different cancers.
Another important strategy to overcome the obstacles of cancer preventive therapy, and to provide a major means to advance the field of cancer chemoprevention, involves the deployment of local therapy. In this regard, it is important to consider several relevant factors. First, the disease we are attempting to modulate is localized. That is, it is at-risk epithelial tissue. Such tissue harbors early changes associated with initial stages of carcinogenesis, and can be broadly characterized into genetic, molecular and physical changes within the epithelium. The epithelium, and the organs which harbor it, are by definition, exposed to the outside environment. In fact, there is ample evidence to indicate that the nature of this environmental exposure serves to aid the initiation and promotion of carcinogenesis by its effects on epithelial and the contiguous stromal cells (6). Of high importance, the major causes of cancer, and of cancer-related death, arise from epithelial tissue. The top four major causes of cancer arise from the epithelial tissue of the breast, prostate, lung, and colon. It is possible to use local delivery methods to selectively deliver therapeutic agents directly to at risk epithelial tissue, and in most cases, this is well within practical means. This strategy either completely avoids, or greatly minimizes, systemic delivery and its resultant toxicity. Furthermore, it delivers high concentrations of the therapeutic agent to the anatomic site at which it is needed. This strategy has the potential to transform our ability to prevent cancer. It is currently under-recognized, underutilized, and under-investigated.
The purpose of this article is to review the rationale of local therapy, what has already been achieved, what studies are ongoing, what is possible, and to thereby provide justification for aggressive pursuit of investigations to expand this strategy which is poised to have a high impact potential for cancer chemoprevention. In particular, advances in the application of this approach are discussed for the prevention of cancer of the skin, breast, lung, colon and rectum, head and neck, esophagus, and bladder. Other organs, such as the kidney, pancreas and ovaries, which are not readily accessible to local delivery agents, are not included in this review.
Pharmacology
Overview
To increase the likelihood of therapeutic success and reduce the occurrence of hazardous drug reactions, a comprehensive understanding of a drug's pharmacologic properties is essential. These properties are divided into pharmacodynamics and pharmacokinetics. Pharmacodynamics refers to the effects exerted by the drug within the body and may include changes at the molecular, cellular, or broader physiologic levels. Pharmacokinetics relates to how a drug moves through the body, and includes how it is absorbed (A) into, distributed (D) throughout, metabolized (M) in, and excreted (E) from the body; together termed a drug's ADME profile. Therefore, pharmacokinetics is how a drug gets there, pharmacodynamics is what the drug does, and together they are termed a drug's pharmacokinetic/pharmacodynamic profile (7).
Drug absorption constitutes the movement of a drug into the bloodstream. The fraction of an orally administered drug that reaches the blood, as compared with when a drug is intravenously administered, is known as its bioavailability. Different routes of administration affect bioavailability for a given drug, for example, in topical (i.e., transdermal) versus oral routes. For a given route of administration, bioavailability values also differ between different drugs. In general, once a drug is present in the bloodstream, it will then have ready access to most organs of the body. Following absorption into the systemic circulation, a drug is distributed to the body's tissues. Tissue distribution is dependent on the partitioning of a drug between the blood and a particular tissue. Some drugs may selectively accumulate in certain types of tissues, such as lipid-soluble drugs tend to selectively accumulate in fatty tissues, while others may be selectively excluded, such as when cancer cells that have developed resistance will pump out certain types of chemotherapy agents. However, this process of distribution typically cannot be closely predicted nor regulated, and in general, most drugs penetrate into a wide variety of different tissue types (i.e., organs) other than the target tissue. The induction of drug effect in nontarget tissues underlies the cause of most drug toxicity.
As soon as a drug enters the body, it can be chemically altered by metabolism. The liver and the associated cytochrome P450 pathway constitute major mediators of metabolism. While some drugs can be activated through metabolic chemical alteration, more commonly, metabolism-associated chemical changes serve to render a drug inactive. Notably, drug metabolism rates can vary widely among patients and may be influenced by coexisting medical conditions, drug interactions, and genetics.
To prevent accumulation of toxic metabolites, inactivated drug compounds and their metabolites need to be removed from the body via excretion. The kidneys are the primary sites for excreting water-soluble compounds. In contrast, larger metabolites that are fat-soluble are more likely to be excreted in bile, after being processed within the liver, and are then eliminated through the gut in feces. Other organs, such as the lungs and skin, also play roles in drug excretion, but they tend to relate to more specialized circumstances. Importantly, pharmacokinetics is a highly dynamic process. As soon as drug begins absorption into the body, the processes of distribution, metabolism, and excretion are in play.
Clinical pharmacokinetics
A drug's overall effect is determined by the relationship between the accessible concentration of the drug and its pharmacologic action upon the target at a given concentration. Clinical pharmacokinetics attempts to quantify the relationship between a drug's dose and subsequent pharmacologic effect. This provides a framework within which to interpret measurements of drug concentration and make the necessary dosing adjustments to achieve clinical efficacy while limiting toxicity. This requires achieving drug concentrations within a therapeutic window, that is, drug plasma concentration above a minimum effective concentration (MEC) but below a minimum toxic concentration (MTC). Determining when a drug reaches the MEC, when it reaches peak concentration and the duration over which the drug concentration exceeds the MEC identifies the duration of a drug's therapeutic effect. Exceeding the MTC corresponds with the onset of adverse effects, signifying drug-related toxicity.
Clearance becomes a very significant concept when designing a regimen that is intended for long-term administration, as the goal is to maintain a steady-state concentration of a drug within a therapeutic window, whereby a drug is clinically effective but below the threshold for drug-related toxicity. Assuming complete bioavailability, steady-state concentration of a drug is achieved when the rate of drug administration is equal to the rate of drug elimination. The bioavailable fraction of a drug that is capable of producing a biologic effect is measured by the area under the curve (AUC) in a plot of drug concentration against time. Thus, AUC represents the most reliable measure of bioavailability, as it is proportional to the amount of drug that is absorbed and represents the total drug exposure over time.
Local delivery
Gastrointestinal pharmacology.
The vast majority of drugs are taken orally to induce a systemic effect and are primarily given as tablets or capsules mainly due to convenience, cost, stability, and patient acceptance. When taken underneath the tongue (sublingual) or in between the gums and cheek (buccal), rapid absorption of a drug can occur due to the thin epithelium and rich vascular supply of the oral mucosa. However, absorption of an orally administered drug may not be uniform. An oral drug encounters several factors that frequently vary between and even within individuals that affects its absorption, including differences in luminal pH, surface area, blood perfusion, presence of bile and mucous, and the nature of the epithelial membranes (8).
Shortly after swallowing, a drug is delivered to the stomach where it encounters a low pH and numerous gastric secretions which may potentially lead to premature degradation. Enteric coating that is stable at the highly acidic pH found in the stomach can be applied to oral medications and helps protect them from disintegrating in the gastric environment. Nevertheless, little absorption occurs in the stomach due to the thick mucosal lining and short transit time. Stomach contents are emptied into the small intestine where numerous permeable microvilli provide the largest surface area in the gastrointestinal tract to facilitate drug absorption. Conditions (e.g., gastroparesis) or medications (e.g., anticholinergic medications) that delay gastric emptying can therefore potentially affect drug absorption. As a drug needs to be dissolved before it can be absorbed, the dissolution rate determines the amount of drug that can be absorbed. Here, intraluminal pH plays a large part in the rate of dissolution as it can vary greatly across the gastrointestinal tract, from a pH of approximately 3 in the stomach to a pH of approximately 7–9 in the small intestine, as well as fluctuating between the fasting and fed state (9).
After absorption through the small intestine, a drug enters into the portal venous system where it is directed to the liver and is initially metabolized before reaching the systemic circulation in what is commonly known as the first-pass effect. This early step in metabolism can reduce the amount of active drug, but also, in some cases, may act to activate a prodrug. Consequently, drugs with a high first-pass effect may require considerably larger doses than parenteral or sublingual doses to achieve adequate systemic concentration, as these alternative routes of administration bypass first-pass metabolism. In addition, individual variability can lead to differences in the degree of first-pass metabolism. For instance, bioavailability can be increased in patients with liver cirrhosis due to ineffective hepatic metabolism (10).
In contrast with traditional drugs that are designed to be absorbed, if only topical application is desired, nonabsorbable formulations can be designed that act within the intestinal lumen and are confined to the gastrointestinal tract without reaching the systemic circulation. Permeability through the intestinal tract is determined by a combination of parallel transport processes. At the level of the enterocyte, diffusion of a drug occurs through paracellular passive transport through the tight junctions as well as active and passive transport, and efflux mechanisms within the enterocytes (11). Drugs with a high molecular weight (i.e., greater than 500 Daltons) and increased polarity are less likely to cross the enterocyte membrane (12). With this in mind, drug design and development may produce formulations that are nonabsorbable and are capable of exerting a therapeutic effect primarily limited to the intestinal tract.
Dermatological pharmacology.
In general, cutaneous therapies can be applied for two main principles: treating local skin disorders or as a route to achieve systemic drug delivery. The chemical nature of the particular drug and its formulation constitute central determinants of whether a drug remains localized to the skin or is absorbed systemically. Not all drugs can readily penetrate the intact skin and understanding the principles that govern percutaneous drug absorption and metabolism through this highly complex structure are essential for safe and effective use. Absorption is determined by the surface area and the nature of the site over which the therapy is applied, along with the lipid solubility of the drug that needs to cross the epidermis, which behaves as a lipid barrier. Thus, highly lipophilic drugs have a greater tendency to be absorbed through the dermis (13).
A drug applied to the skin must first pass through a layer of dead epidermal cells, known as the stratum corneum (the outer 5–600 μm), which acts as a barrier against infection, dehydration, and mechanical stress. Through multiple proteins and lipids that may bind to drugs, the stratum corneum can function as a drug reservoir, allowing a drug to diffuse into the skin even after topical application of the drug has ceased. Conditions that compromise the body's ability to maintain this barrier (e.g., burned or denuded skin) or are associated with increased cutaneous blood flow (e.g., inflammatory skin disorders), can considerably enhance drug absorption. In terms of transdermal delivery, where the drug is intended to penetrate the dermis and enter the local or systemic circulation, vehicles such as oils and ointments can facilitate drug permeation and enhance the efficacy of topical preparations.
Pulmonary pharmacology.
Inhalation is an established route of administering a drug that enables direct delivery to the respiratory tract, thereby providing for relative selectivity for delivery to lung tissue. It is the preferred route for drug delivery in acute respiratory emergencies, particularly for asthma and COPD (14), when administering β2 agonists, anticholinergic medications and corticosteroids, but also for antibiotics for chronic respiratory infections (e.g., cystic fibrosis), or prophylaxis (e.g., against Pneumocystis carinii in immunocompromised patients).
Particle size and the effective delivery of an inhaled therapeutic through a respiratory device can determine the extent of drug distribution between the lungs and oropharynx (15). Whereas, larger particles may settle in the upper airways, the optimal size for a particle to settle in the smaller airways is 2–5 μm mass median aerodynamic diameter (MMAD). This is critical to understand with management of chronic obstructive pulmonary disease (COPD) or severe asthma, where a high degree of luminal narrowing serves to limit airflow, and is made ever worse by associated mucous and cellular exudate, which further compounds occlusion, all serving to limit drug deposition into the small airways (16).
In addition, the majority of a drug (80%–90%) that is inhaled through a conventional pressurized metered-dose inhaler (pMDI) and enters the oropharynx is swallowed, absorbed, and enters the systemic circulation, after undergoing first-pass metabolism through the liver (17). Therefore, only 10%–20% of a drug will be directly delivered to the lungs, and this too has the potential to be absorbed into the systemic circulation. Methods that optimize delivery to the smaller airways include drugs formulated in hydrofluoroalkane (HFA) particle propellant; this maximizes drug deposition to the lower respiratory tract when used with respiratory devices (18). Pressurized metered-dose inhalers are in common use, due to cost, convenience, and portability, and their ability to deliver 100–400 doses of a drug. However, synchronizing the activation of a respiratory device and inhalation can be difficult and is a major cause of treatment failure despite appropriate instruction.
With the aid of large-volume spacer chambers secured with one-way valves between the pMDI and the patient, evaporation of a liquid propellant helps reduce a particle's velocity and size, thus, reducing the need for carefully coordinated breaths for appropriate drug delivery and enhancing the amount of drug that deposits in the lower airways, rather than incorrectly in the oropharynx. In acute COPD and asthma exacerbations where there is severe airway obstruction, nebulized drugs, through either a jet nebulizer driven by a stream of gas (air or oxygen) or an ultrasonic nebulizer, offer a more efficient method of delivering large doses during tidal breathing.
Skin Cancer
Background: skin cancer
The skin constitutes the largest organ of the body. It is also the source of the greatest number of cancers and precancerous lesions. In fact, this number is so high that it is common practice to exclude non-melanomatous skin cancer when discussing the incidence of other cancers. It is therefore notable that skin is directly exposed to the environment, is readily accessible anatomically, and is highly amenable to the delivery of localized therapy.
There are three main types of skin cancer: basal cell carcinoma (BCC), squamous cell carcinoma (SCC), and melanoma. Exposure to ultraviolet (UV) radiation is the most preventable risk factor for all the three main types of skin cancer. Of the three wavelength ranges that are emitted from the sun in the UV-spectrum, UVA and UVB represent 95% and 5% of the UV radiation (UVR) that is able to penetrate the atmosphere and reach the earth's surface. UVA (320–400 nm) passes through the epidermis and into the dermis, whereas UVB (280–320 nm) enters into the epidermis, but incompletely extends to the dermis. There are many oncogenic mechanisms by which radiation has been implicated in skin carcinogenesis. A main one relates to UVR causing DNA damage and genetic mutations (e.g., p53 mutations seen in 45%–60% of SCCs), which are initiating events in skin carcinogenesis (19). Second, UV-induced immunosuppression may interfere with the ability of the immune system to identify and eliminate cancer cells before they become clinically apparent skin tumors. UVR may also generate reactive oxygen species, which separately lead to DNA damage, lipid peroxidation, and activation of signal transduction pathways.
For skin cancer prevention, the established mainstay of the approach involves local measures aimed at exposure and early detection. By virtue of their frequency and accessibility to preneoplastic intermediates amenable to local therapy, skin cancers represent one of the more preventable cancers. As cumulative UV exposure from natural and artificial sources correlates strongly with the development of all types of skin cancer, behavioral interventions aimed at avoiding UV exposure are strongly emphasized as the primary method to reduce skin cancer incidence. There are several methods to limit sun exposure to UVR including: sun avoidance (not going outside during peak UVR times), wearing protective clothing, and educating the public against the use of tanning salons. The application of topical preparations represents another method, and it constitutes an important manner by which effective therapeutics can be delivered via a local delivery route. Sunscreens behave as a protective barrier to the skin to prevent UV-related carcinogenesis.
Approach to prevention: topical therapy
Excluding total sun avoidance, sunscreens represent the most effective and readily available strategy for photoprotection from UV-induced skin damage. Of UVR, UVB radiation is recognized as the most likely to be erythrogenic and responsible for photocarcinogenesis. UVA radiation, which previously was thought to only cause cosmetic damage, may also contribute to the development of skin cancers. Ideal sunscreen preparations should cover a broad spectrum, providing appropriate coverage for both UVA and UVB radiation. The sun protection factor (SPF) measures a sunscreen's ability to protect against UVB but does not reliably measure protection against UVA radiation. The American Association of Dermatology recommends everyone choose a sunscreen that is broad-spectrum, has a SPF of 30 or higher, and is water-resistant (20). Many of the sunscreens available are a blend of two major active ingredients: (i) chemical agents that absorb solar radiation in the UV range and convert the energy to heat or light and (ii) physical agents that block or reflect ultraviolet radiation and reduce its transmission to skin (13).Currently available UVA filters in the United States include avobenzone, oxybenzone, titanium dioxide, zinc oxide, and ecamsule. There are numerous UVB filters, including p-aminobenzoic acid (PABA), cinnamates, octocrylene, and salicylates.
Regular application of sunscreen can reduce the risk of cutaneous SCCs and its precursor lesion, actinic keratosis (21). In one community-based trial following 1,621 residents in Queensland, Australia, daily application of SPF 15-plus sunscreen was associated with a statistically significant 46% decrease in the incidence of SCCs compared with placebo, although no effect was seen with overall BCC incidence (22). Even after 8 years following this trial, the rates of SCCs were decreased by nearly 40%, signifying a prolonged preventative effect (23). However, despite a decreased trend, there remained no significant difference observed in the incidence of BCCs. Further follow up at 10 years after trial completion revealed a 50% reduction in the incidence of melanoma cases, and over 70% reduction in invasive melanoma, in the sunscreen intervention group (24).
With a relative risk nearly 100 times higher than an immunocompetent population, organ transplant patients have an exceptionally high rate of developing actinic keratosis and SCC (25). In a prospective, case–control trial, 120 solid organ transplant patients were followed for 24 months to evaluate the benefit of daily sunscreen use versus intermittent use, in the context of an overall UV protection strategy. Participants using daily sunscreen had fewer actinic keratosis and SCCs, and to a lesser extent BCCs, than the participants using sunscreen intermittently.
As a universally available medication that has an excellent safety profile and negligible systemic absorption, sunscreen is widely supported for skin cancer prevention by many public health societies, yet, it remains far from a regular practice among the general public. There are currently no published studies that indicate routine use of sunscreen is hazardous to humans and the scientific consensus remains supportive of its use among the general public. Other than local benign irritant, allergic reactions, and the inconvenience of application, sunscreens are generally well tolerated and have low systemic absorption. Studies have called into question the safety of particular sunscreens with the uptake of dissolved organic UV filters detectable in blood (∼0.1%–5%; ref. 26) and urine after topical application. Insoluble UV filters, such as titanium dioxide, do not penetrate through the outer layer of human skin and pose no potential risk for systemic toxicity (26).There is currently no evidence that systemic absorption may result in biological consequences. For example, causal associations to thyroid and sex hormone disruptions attributed to systemic absorption of UV filters have been rejected by confirmation studies that have been unable to identify any disturbances to the endocrine system (27, 28).
Another area of concern with sunscreens relates to its potential interference with vitamin D synthesis. Exposure to sunlight is essential in the first step in the synthesis of the biologically active form of vitamin D (1,25-dihydroxy-vitmain D3). As this solar-induced process is responsible for the production of 90% of vitamin D in most people, interrupting this mechanism by blocking UVB penetration has raised concerns for possible vitamin D deficiency (29). However, prospective and retrospective studies examining this relationship have been unable to correlate frequent sunscreen use with vitamin D deficiency (30, 31). Nevertheless, oral Vitamin D supplementation would be the preferred alternative to the risk of unprotected exposure to UV radiation.
Despite the universal availability of sunscreens, excellent safety profile and insignificant systemic absorption, treatment patterns indicate that it remains far from a regular practice among the general public. In a report published by the Centers for Disease Control and Prevention in 2010, 31% of U.S. adults reported applying sunscreen with SPF of 15 or higher (32) and even lower usage rates are reported among U.S. teens (33). Similar rates are observed in a population study in northern England with only 35% females and 8% males reporting regular use of sunscreen (34). High costs of sunscreen products, low cosmetic appeal (35), demographic disparities, improper application of sunscreen, questions about sunscreen durability after water exposure with required readministration to maintain adequate protection, and the behavioral changes necessary to apply sunscreen, all represent reasons that serve to limit increased sunscreen utilization by the general public (26).
Actinic keratosis arises from intraepidermal proliferation of atypical keratinocytes in response to cumulative exposure to the UV radiation. These lesions, which affect 5.5% of women and 13.9% of men in the United States as well as up to 40%–60% of Australian adults (36), represent an intermediate in the sequence of progression from normal skin to invasive cancer, and as such, they are considered a preneoplastic lesion that precedes the development of in situ cancer. Although the majority of actinic keratosis cases do not progress to SCCs, 65%–72% of cutaneous SCCs arise in association with actinic keratosis (37). The relationship between actinic keratosis and BCCs is less clear. No distinguishable factors can predict those at higher risk for the development of invasive lesions, and therefore the majority of patients pursue treatment to relieve concerns about potential future cosmetic sequelae and/or receiving a diagnosis of invasive cancer.
The therapeutic options for actinic keratosis primarily depend on the number of lesions and the efficacy of treatment (38) and fit into two broad categories: surgical and ablative therapies to individual lesions, or medical therapy. An individual lesion can be treated with liquid nitrogen cryosurgery, which can be quickly performed in an office-based setting, is well tolerated by patients, and has a cure rate of 99% after one year of follow-up (39). Medical therapy is particularly effective in treating large areas with multiple actinic keratosis and can be treated with five currently FDA-approved targeted topical therapies: 5-fluorouracil cream (5-FU), 3% diclofenac gel in 2.5% hyaluronic acid (hyaluronic acid limits diclofenac absorption), imiquimod cream, ingenol mebutate gel, and δ-aminolevulinic acid with photodynamic therapy.
Evidence derived from multiple randomized trials and systematic reviews have found similar efficacy rates between these field-directed therapies in treating actinic keratosis, with complete response seen in nearly half of these patients (40, 41). The lack of direct comparison trials makes it difficult to determine which of these modalities is the most efficacious. The most common adverse effects included local site reactions, erythema, scaling, crusting, pruritus, and facial irritation. With minimal systemic absorption (calculated rates as high as 2.4% with 5-FU 5% topical cream; ref. 42) serious adverse reactions are not common. Results from selected trials are depicted in Table 3. Although no clinical trials with the prevention of SCC as a primary endpoint have been conducted, it is assumed that interventions for actinic keratosis would translate into a preventative effect leading to a reduction in SCC incidence. Therefore, response rates in actinic keratosis have been interpreted as an acceptable surrogate marker for clinical benefit.
Selected trials of FDA-approved agents for actinic keratosis treatment
Intervention . | Studiesa . | Participant complete clearance ratio (%) . | Efficacy rate . | Common adverse reactions . | Systemic absorption . |
---|---|---|---|---|---|
5-FU | Jorizzo 2002 | 56% | 52% (196) | Inflammation, crusting, temporarily disfiguring | 0.55–2.4% (42) |
Weiss 2002 | 48% | ||||
3% diclofenac gel in 2.5% hyaluronic acid | Wolf 2001 | 45% | 40% (197) | Dry skin, pruritus, erythema | 0.6–2.2% (198) |
Rivers 2002 | 31% | ||||
Gebauer 2003 | 38% | ||||
Imiquimod | Chen 2003 | 24% | 50% (199) | Erythema, scabbing, flaking | Minimal |
Lebowhl 2004 | 45% | ||||
Korman 2005 | 48% | ||||
Ooi 2006 | 45% | ||||
Alomar 2007 | 55% | ||||
Jorizzo 2007 | 54% | ||||
Ingenol mebutate | Anderson 2009 | 54% | 42% (200) | Erythema, scaling, crusting, edema, ulceration, pruritus | Minimal |
Lebowhl 2012 | 42% | ||||
δ-aminolevulinic acid + PDT | Hauschild 2009 AK03 | 62% | 42%–89% (201) | Burning sensation during light exposure, irritation | Minimal |
Szeimies 2010 | 60% |
Intervention . | Studiesa . | Participant complete clearance ratio (%) . | Efficacy rate . | Common adverse reactions . | Systemic absorption . |
---|---|---|---|---|---|
5-FU | Jorizzo 2002 | 56% | 52% (196) | Inflammation, crusting, temporarily disfiguring | 0.55–2.4% (42) |
Weiss 2002 | 48% | ||||
3% diclofenac gel in 2.5% hyaluronic acid | Wolf 2001 | 45% | 40% (197) | Dry skin, pruritus, erythema | 0.6–2.2% (198) |
Rivers 2002 | 31% | ||||
Gebauer 2003 | 38% | ||||
Imiquimod | Chen 2003 | 24% | 50% (199) | Erythema, scabbing, flaking | Minimal |
Lebowhl 2004 | 45% | ||||
Korman 2005 | 48% | ||||
Ooi 2006 | 45% | ||||
Alomar 2007 | 55% | ||||
Jorizzo 2007 | 54% | ||||
Ingenol mebutate | Anderson 2009 | 54% | 42% (200) | Erythema, scaling, crusting, edema, ulceration, pruritus | Minimal |
Lebowhl 2012 | 42% | ||||
δ-aminolevulinic acid + PDT | Hauschild 2009 AK03 | 62% | 42%–89% (201) | Burning sensation during light exposure, irritation | Minimal |
Szeimies 2010 | 60% |
aStudies selected by: randomized, double-blinded controlled trial, complete clearance ratios for lesions listed, toxicity grade standardized, included both head and non-head regions, non-immunocompromised.
Breast Cancer
Background: breast cancer
In 2016, there will be an estimated 250,000 new cases and 40,000 deaths attributed to breast cancer in the United States, making it the most commonly diagnosed cancer and the second most common cause of cancer-related death in women in the United States, eclipsed only by lung cancer (43). The burden of disease is greater in less developed countries, as it remains the leading cause of cancer-related death among females (44).
There is a consistent relationship supporting a hormonal basis to breast cancer risk. High endogenous estrogen levels increase the risk of breast cancer (particularly hormone receptor–positive breast cancer) in both postmenopausal (45) and premenopausal (46) women. Likewise, increased progesterone (P4) and progestin exposure is increasingly recognized as an independent risk factor for breast cancer. Clinical findings from the Women's Health Initiative (WHI) illustrate the role of progesterone in the development of breast cancer. Women taking combination hormone therapy with progestin and estrogen experienced a greater breast cancer risk relative to estrogen alone and were susceptible to larger and higher grade neoplasms (47). Similarly, oral contraceptive use has been associated with a modest increase in breast cancer risk in premenopausal women (48) but more recent data show that the use of depo-medroxyprogesterone acetate (DMPA) carries a 2-fold increase in breast cancer risk, with a trend toward higher grade and triple-negative tumors (49).
Hormone therapy targets hormone receptors on hormone-sensitive breast cancer cells and acts to prevent the development of breast cancer. Primary prevention in breast cancer ranges from selecting high-risk patients as candidates for endocrine therapy to prophylactic mastectomies in women with the inherited germline mutations BRCA1 and BRCA2 who have an estimated 50% risk of breast cancer in their lifetime. Coincident with discontinuation of exogenous hormone replacement therapy (HRT), improved therapeutic agents and a major public health focus on breast cancer prevention, breast cancer–related death rates have followed a downward trend since 1989 in the United States (50).
Currently available endocrine therapies for women at high risk for breast cancer include selective estrogen receptor modulators (SERM), tamoxifen and raloxifene, and aromatase inhibitors (AIs; anastrozole and exemestane), the choice depending on their menopausal status and patient preference. Although, only tamoxifen and raloxifene are approved for breast cancer prevention in women at higher risk. In 1998, the National Surgical Adjuvant Breast and Bowel Project, Prevention-1 (NSAB P-1) demonstrated that daily tamoxifen for 5 years reduced the risk of invasive breast cancer and noninvasive breast cancer by half, with the most significant reduction being seen in estrogen receptor (ER)-positive invasive breast cancer, which was reduced by 69% (1). This was followed by the Study of Tamoxifen and Raloxifene (STAR), which identified another SERM, raloxifene, as equally effective in reducing the risk of invasive breast cancer as tamoxifen, but observing a lower incidence of thromboembolic events and cataracts (51). For women with ductal carcinoma in situ (DCIS) treated with breast-conserving therapy and radiotherapy, postoperative tamoxifen for 5 years reduces the risk of breast cancer recurrence, although it has no effect on survival (52). Most recently, the Adjuvant Tamoxifen: Longer Against Shorter (ATLAS) trial, revealed 10 years of adjuvant tamoxifen reduced breast cancer recurrence and a small but statistically significant reduction in all-cause mortality compared with 5 years of therapy (21.5% vs. 22.9%) at the expense of slightly increased rates of endometrial cancer and thromboembolic disease (53).
Studies of aromatase inhibitors in postmenopausal women have also been shown to prevent invasive breast cancer. This was seen in the International Breast cancer Intervention Study (IBIS-II) that showed a 50% reduction in invasive breast cancer and DCIS with anastrozole compared with placebo (54). In a separate placebo-controlled trial, at 36 months of follow-up, postmenopausal women who took exemestane were 65% less likely to develop breast cancer than those who took placebo (55). These studies examining aromatase inhibitors, however, were not powered to show a survival benefit. Overall, the data suggest that aromatase inhibitors are reasonable alternatives to SERMs. While no aromatase inhibitors are currently approved for breast cancer prevention in the United States, ASCO clinical practice guidelines (56) recognize exemestane as a rational substitute to SERMs in postmenopausal women who desire breast cancer prevention.
The demonstration that breast cancer can be prevented in high-risk women through therapeutic intervention constitutes a major achievement for the field of cancer chemoprevention. However, despite this achievement, there has been widespread reluctance to adopt these strategies for breast cancer prevention. The central reason for this reluctance relates to side effects of therapy. There are several relevant side effects. Tamoxifen and raloxifene are both associated with an increased risk of thromboembolic events. Tamoxifen, the first FDA-approved agent for breast cancer prevention, increases the risk of endometrial cancer. Although the overall incidence of endometrial cancer was very low, could be detected at a very early stage that was highly amenable to effective therapy and had no overall impact on clinical outcome, the procarcinogenic effect by an agent designed to inhibit breast cancer carcinogenesis was not well received by neither the general medical community nor the general public. In addition, tamoxifen exerts several additional effects including thromboembolic events, vasomotor instability, clinically manifested as flushing, as well as an increased incidence of cataracts. Similarly, aromatase inhibitor use is associated with an increased risk of musculoskeletal side effects, hypertension, vasomotor symptoms, and loss of bone density, which is an important consideration in this population. In addition, aromatase inhibitors are contraindicated in premenopausal women with intact ovarian function. Furthermore, the reduction in breast cancer risk from both SERMs and aromatase inhibitors in primary prevention did not translate into a reduction in breast cancer–specific or all-cause mortality (54, 57). Taken together, the general consensus was that the toxicity profile of oral agents outweighed their benefit related to breast cancer prevention.
When tested in postmenopausal women with relapsed PR-positive metastatic disease who had received prior adjuvant hormone/chemotherapy, mifepristone produced only a partial response in approximately 10% of patients, and was not considered a successful outcome (58). Among the newer generation of SPRMs, namely telapristone and ulipristal, recent data suggest a substantial tumor-protective effect through PR blockade that is more specific than mifepristone. With potent anti-progesterone effects and less antiglucocorticoid activity than mifepristone, these SPRMs possess a more favorable toxicity profile.
Approach to prevention: local transdermal therapy
Given that it is possible to prevent breast cancer through pharmacologic intervention, but that intervention is limited by systemic toxicity, it would therefore be highly desirable to deliver therapy locally. Intraductal injection was previously proposed as a novel method to deliver local therapy, as it demonstrated superior efficacy compared with intravenous injection in animals (59). However, this costly and invasive procedure requires significant expertise in identifying a single duct for injection of therapeutic agents and has not been tested in humans. Furthermore, this approach has significant practical limitations when one considers that each duct of the breast must be individually accessed.
The breast is ideally suited to delivery of therapeutic agents through a local transdermal therapy (LTT) strategy. Drugs that are highly lipophilic and have a low molecular weight (<500 Daltons) have the potential to effectively permeate the stratum corneum. Among the drugs that have been successful in crossing this barrier percutaneously are the reproductive hormones, including those used in hormonal replacement therapy (60). Typically, drugs applied to the skin on other regions of the body penetrate the skin and are absorbed into the systemic circulation. Importantly, embryologically, the breast arises from the skin and is thus a modified skin appendage. As such, the breast has its own well-developed internal lymphatic circulation that facilitates dissemination throughout the breast. Therefore, with LTT, a drug can be delivered to the breast, disseminate throughout the breast through the internal lymphatic circulation, provide high drug concentrations within the breast for little to no systemic penetration, and thereby, produce little to no systemic side effects (61). Furthermore, the relatively thin epidermis of the mammary papilla and multiple duct openings on its surface that directly extend into the terminal duct lobular units of the breast creates a situation that enables efficient transdermal delivery (62).
Tamoxifen undergoes extensive metabolic conversion, including by the cytochrome P450 system (e.g., CYP2D6 and CYP3A4/5 isoforms) to its major antiestrogenic metabolites: 4-hydroxytamoxifen (4-OHT) and N-desmethyl-4-hydroxytamoxifen (endoxifen; ref. 63). These active metabolites have 25–100 times the affinity to the ER-α compared with their parent drug. Women with certain genetic CYP2D6 variants poorly metabolize tamoxifen, leading to reduced production of endoxifen, consequently increasing the risk for breast cancer recurrence. Screening for this genetic mutation prior to oral tamoxifen remains controversial, and is not routinely performed (64). Delivery of 4-OHT or endoxifen to target tissue would bypass the liver, would circumvent the impact of genetic variations in metabolism, and would thereby negate the need for testing while ensuring uniform delivery of effective therapy.
4-OHT is a small molecule whose chemical properties make it a suitable candidate for LTT. Pharmacokinetic studies have shown that 4-OHT is able to effectively penetrate the skin and selectively concentrate in breast tissue (65, 66). Its high potential as a chemopreventive agent after local delivery has recently been demonstrated in a phase II trial conducted by our group (67). In that study, women with DCIS were randomly assigned to treatment with oral tamoxifen or with topical 4-OHT, in a double-blinded, placebo-controlled fashion, in the window prior to surgical intervention. In this manner, it was shown that topical 4-OHT led to equivalent tissue concentrations of active agent, compared with oral tamoxifen, as well as equivalent and significant decreases in cell proliferation. The latter was measured by Ki67 staining, comparing posttreatment biopsy samples to those of pretreatment.
Of high importance, there was essentially no systemic absorption of topical 4-OHT, nor was there any evidence of systemic toxicity. Specifically, with topical 4-OHT, the blood concentrations of active species were minimal-to-none, and were far below those seen with oral tamoxifen. There are several active metabolites of tamoxifen, and this lack of systemic distribution was observed for several major and pharmacologically relevant metabolites. By avoiding metabolism of oral tamoxifen in the liver, transdermal 4-OHT avoids variations in drug concentration due to genetic variations in CYP450-based metabolizing enzymes. Furthermore, systemic biomarkers of tamoxifen-mediated hypercoagulation, Von Willebrand factor and factor VIII, were significantly elevated in those receiving oral tamoxifen, but not in women receiving topical 4-OHT (63, 64). Finally, systemic biomarkers of tamoxifen-associated hormonal action included an increase in sex hormone–binding globulin (SHBG) and a decrease in insulin-like growth factor-1 (IGF1), and were significantly altered by oral tamoxifen but not by topical 4-OHT (65, 68).
Therefore, while oral tamoxifen led to high systemic concentrations of drug and associated systemic effects associated with tamoxifen-induced toxicity, local delivery of 4-OHT was equally efficacious in a targeted fashion on high-risk breast tissue, but had no systemic consequences. The findings of this study were very important for several reasons. They demonstrated for the first time the proof-of-principle concept that internal organs at risk for the development of cancer could be pharmacologically targeted with cancer chemopreventive agents whose mechanism of action serves to disrupt the process of carcinogenesis. Furthermore, it demonstrated that such targeting induced biologically relevant efforts on high-risk target tissue, while sparing systemic side effects. Finally, this study focused on breast cancer, whose high prevalence constitutes a major public health problem.
The positive findings with the 4-OHT justify future investigations related to 4-OHT specifically and more broadly with the concept of chemoprevention using a targeted local delivery strategy in breast and other organs. Although the 4-OHT study findings were significant, they were achieved in the context of a single institution small phase II trial. Important next steps include independent repetition and expansion of size. The latter will allow for initial interrogation of whether 4-OHT induces uterine hyperplasia, and examination of effects on vasomotor symptoms. The cohort examined in the completed phase II trial was perimenopausal, was experiencing a relatively high level of such symptoms, and the small sample size did not permit identification of significant differences between baseline, oral tamoxifen, or transdermal 4-OHT.
Findings from the phase II 4-OHT trial demonstrate the feasibility of local delivery of agents to the breast, and thereby provide justification for a broader examination of agents. These are now being actively pursued by us and others. Telapristone is a selective progesterone modulator (SPRM), and we have implemented a chemoprevention trial of LTT to the breast (ClinicalTrials.gov: NCT02314156). As a SPRM, telapristone leads to progesterone receptor blockade leading to changes in gene expression and deceasing progression of a mutated cell through the cell-cycle progression (69). In preclinical testing, topical telapristone has been studied in comparison with systemic telapristone delivered subcutaneously in a multi-arm trial involving athymic nude rats with primary focus on tissue and plasma concentrations (70). Mammary tissue concentrations were significantly higher in those rats receiving topical preparations compared with recipients of systemic telapristone. However, plasma concentrations were similar between recipients in topical and systemic treatment groups likely due to a structural difference rather than a drug trait as rats do not possess a distinctly separate breast structure as do humans. In NCT02314156, women who are undergoing therapeutic or prophylactic mastectomy are randomly treated with oral versus topical telapristone prior to surgery. Endpoints include tissue and blood drug concentrations, and biomarkers of therapeutic efficacy.
Additional agents hold promise, and are at various stages in development. Endoxifen (N-desmethyl-4-hydroxytamoxifen) is an important tamoxifen metabolite whose pharmacologic properties play an important role in mediating tamoxifen's therapeutic effect. In part, this relates to the facts that endoxifen has a serum concentration ten times greater than that of 4-OHT, leads to proteosomic degradation of ERα, inhibits ERα transcriptional activity, and will inhibit estrogen-induced breast cancer cell proliferation (63). Emerging data from the first human prospective trial examining endoxifen's clinical efficacy indicate that it demonstrates potent antitumor activity with acceptable toxicity and has supported its investigation as a potential chemopreventive agent (71). The addition of drug vehicles enhance permeation through the skin and have demonstrated efficient absorption that is compatible with transdermal therapy (72). To date, endoxifen's in vivo benefit has only been studied in clinical trials using oral preparations of the drug but studies using topical preparations are of future research interest.
It is important to consider that LTT to the breast should not necessarily be limited to hormonal agents. In this regard, it is also important to consider that inflammation has been extensively implicated in breast carcinogenesis. The nonselective COX inhibitor, diclofenac, holds promise as a breast cancer prevention agent (73). It was tested in a preoperative setting in women scheduled for mastectomy to prevent or treat breast cancer by comparing tissue and systemic concentrations between patch application to the breast versus application to the abdomen. Transdermal application to the breast provides for selective local delivery to the breast, while application to the skin of the abdomen constitutes a form of systemic delivery. That is, after application to the skin of the abdomen, drug is absorbed into the systemic blood system, whereupon it is delivered to all organs of the body, including the breast. Results from this study affirmed the prediction that topical application to the breast would selectively concentrate drug in the breast parenchyma. However, the authors did stress that accumulation was in a nonhomogenous pattern. No previous studies, including oral preventative agents, have attempted to define drug distribution throughout the entire breast using multiple tissue sampling locations. Therefore, the degree to which this differs from distribution attained after oral administration of an agent is not yet clear. The authors did emphasize that the more relevant endpoint for transdermal therapy is biological effect rather than drug concentration, which calls for larger clinical trials to test for efficacy and safety. In this regard, it is important to consider that distribution of small-molecule therapeutics within the breast, after transdermal delivery, will be mediated by lymphatic flow. Thus, distribution at any given point in time will be affected by physical factors, inclusive of manipulation of the breast. Therefore, real-time measurement of concentration may be misleading, while biological effect will reflect exposure over an extended period of time.
Lung Cancer
Background: lung cancer
As the leading cause of cancer-related mortality facing the world population for both men and women, lung cancer remains one of the most formidable challenges for clinicians and researchers. More people die annually of lung cancer than of colon, breast, and prostate cancer combined (74). Because of the absence of pain sensation in the lung, the presence of symptoms such as cough, hemoptysis, dyspnea, and chest pain often reflects invasion into local or distant structures. This in part explains the fact that approximately three of four lung cancers are at an advanced stage at diagnosis, where potentially curative surgical resection is no longer an option (75). At this stage, conventional approaches such as surgical resection, radiation, and chemotherapy produce a discouraging 15% five-year survival rate (76), prompting the search for additional strategies to reduce the burden of lung cancer.
Smoking is the most clearly documented risk factor (77) associated with the development of lung cancer and a strong emphasis on smoking prevention and cessation remains the key element in any comprehensive preventative strategy. Nevertheless, it has not been possible to substantially increase rates of smoking cessation. Furthermore, the increased risk of developing lung cancer persists many years after smoking cessation. Recent data reveal a downward trend in the rates of smoking among developed countries. However, disturbingly, tobacco products are designed to be addictive and sellers of tobacco products are now targeting countries such in Southeast Asia and China, where rates of smoking are on the rise (78).
For many years, effective early detection strategies had been pursued, and were elusive. The logic being that survival of early-stage lung cancer, including cure, is better than that of more advanced stages. This all changed with the National Lung Screening Trial (NLST). It was a notable achievement as it revealed screening with low-dose CT in high-risk adults, including active and former smokers, reduced mortality from lung cancer (79). On the basis of this study, the American Lung Association and the American College of Physicians/American Society of Clinical Oncologists released screening guidelines recommending annual low-dose spiral CT screening for smokers and former smokers ages 55–74 who have smoked for 30 pack-years or more and either continue to smoke or have quit within the past 15 years (80).
Unfortunately, chemoprevention trials specifically designed to assess effects on cancer development have so far been unsuccessful. Phase III clinical trials including the α-Tocopherol, β-Carotene (ATBC) Study and the β-Carotene and Retinol Efficacy Trial (CARET) in current and former smokers demonstrated vitamin and micronutrient supplementation to be ineffective, and even in some cases, harmful, increasing the risk of death from lung cancer and cardiovascular disease (81, 82).Consistent with these results, the Physicians' Health Study which randomized former smokers to β-carotene and/or aspirin to placebo similarly revealed no benefit (83). Notably, these prospective trials were not founded on compelling preclinical data from animal studies and early-phase clinical trials that provide a strong rationale of efficacy in humans which typically precede large and lengthy definitive phase III clinical trials.
Similar to other solid tumors, a series of alterations in gene expression and chromosome structure occur throughout the development of lung tumorigenesis from normal lung epithelial cells toward hyperplasia, metaplasia, dysplasia, and carcinoma in situ to invasive cancer. As seen with the vitamin studies, the sequence to invasive cancer progresses over many years making it difficult to assign lung cancer incidence or mortality as a clinical endpoint in longitudinal studies. Repeated bronchoscopies for histologic diagnosis with biopsies would be ideal but would be logistically difficult, prohibitively expensive, and take decades to complete for conclusive results. Distinct histologic lesions can be reproducibly graded as precursor lesions correlating with the future risk of developing non–small cell lung cancer (NSCLC; ref. 84) and are the primary focus of many lung cancer prevention trials which focus on resolution or regression of dysplastic lesions and pulmonary nodules as surrogate markers to demonstrate a therapeutic benefit (85–87). Current surrogate endpoints include endobronchial dysplasia, bronchial epithelial proliferation, and the presence of CT-detected pulmonary nodules. None of these endpoints have been independently validated to verify a positive benefit but are rather justified on the basis of biologic plausibility.
Approach to prevention: Inhalational therapy
As an internal organ amenable to local delivery strategies for cancer chemoprevention, the lung is second-to-none in offering several potential advantages. It interfaces with the external environment in a highly robust fashion. Its ongoing exchange with environmental gasses provides for a well-established route of delivery via inhalation of aerosolized drug. The high prevalence of lung cancer and its disastrous clinical impact, coupled to our newfound ability to identify refined high-risk cohorts through the use of low-dose CT scan–based screening, all provide strong rationale for an aggressive pursuit of effective chemoprevention strategies for lung cancer.
For years, aerosolized agents have been utilized to treat local pulmonary diseases such as asthma, COPD, pulmonary arterial hypertension, and various microbial infections. The role of inhalational chemotherapy has been well studied in humans, yet, it is not standard of care and its use remains reserved to research in select cases in the adjuvant setting of advanced cancer (88). Further assessment is still needed to evaluate its safety and efficacy in early-stage cancer or in the preventive setting. Nonetheless, the integration of nanotechnology and pulmonary delivery of drug aerosols has helped steer cutting-edge research into a new and exciting direction, applying this science to the concept of nanoprevention. The primary goal of nanoprevention through drug aerosols is the sustained release of low-dose chemoprotective drugs to target tissue to reverse or terminate the dysplasia to neoplasia sequence without any compromising systemic effects.
The upper airway and lungs are in direct contact with the environment and continuously interface with the atmosphere providing a direct route for drug administration. With a large alveolar surface area (near 100 m2), very thin air–blood barrier (<1 μm), extensive vascularization, rapid absorption, lack of first-pass metabolism, high bioavailability, and the capacity to absorb large quantities of drug, drug delivery through the pulmonary route provides an optimal method of directly administering chemopreventive agents to the lungs intended for local or systemic therapy (89).
As adenocarcinomas typically present in the lung periphery and squamous cell carcinomas have a tendency to arise from the central portions of the lung, understanding a therapy's physical characteristics can aid in the development of a pulmonary chemoprevention strategy that attempts to direct therapy to target precancerous lesions arising from different lung regions. As such, a number of micrometer- and nanometer-sized drug carrier systems including liposomes, polymer conjugates, polymeric micelles, microparticles, and nanoparticles have been engineered to selectively deliver various preparations to the lung. In the case of aerosol delivery, after inhaled nanoparticles are dispersed throughout the alveolar sacs, biodistribution and systemic absorption is largely dependent on nanoparticle size and surface charge (90). While, neutrally charged nanoparticles smaller than 6 nm can readily enter the blood stream via transcellular processes, noncationic nanoparticles larger than 34 nm are engulfed and processed by alveolar macrophages. The aerodynamic behavior of a particle as it travels through the airways can be improved through a process known as controlled crystallization that increases a drug's respirable fraction and deposition into peripheral areas of the lung (91).
Nevertheless, there are limitations to inhalation therapy that need to be considered. The lungs have respiratory defense mechanisms adept to protecting the lung epithelium from foreign bodies/particulates, including beating cilia, mucous production, and macrophages, which may render respiratory agents ineffective. Likewise, underlying respiratory diseases, including reactive airway disease (COPD, asthma) and cystic fibrosis that are classically associated with local bronchoconstriction and frequent inflammation may prevent the drug from properly depositing into the target regions. Atelectasis, tumor mass, or pleural effusions are other obstacles to effective drug distribution. Local effects vary from benign and easily treatable, such as cough and oral candidiasis, to more problematic, such as damage to lung parenchyma and exacerbations triggered by bronchoconstriction. As long as toxicity and side effects are minimized, inhalation therapy can provide a highly efficient mode of drug delivery for preventative purposes.
The role of pulmonary inflammation induced by tobacco smoke has been directly linked with the development of lung cancer. Inhaled glucocorticoids, known for their anti-inflammatory properties, have been an enduring and integral component of standard management for asthma and advanced COPD. In murine lung carcinogenesis models, inhaled corticosteroids effectively inhibited all stages of progression from hyperplasia to the formation of cancer (92) and delayed the appearance of lung tumors and decreased their growth and progression to carcinomas (93). In an analysis of 10,000 patients at Veterans Affairs Medical Centers (VAMC) who received inhaled corticosteroids for COPD and were highly compliant, there was a significantly reduced risk of lung cancer (86).
Predicated on this preclinical and epidemiologic evidence that supported a positive benefit of aerosolized steroid use on lung cancer incidence, several phase II clinical trials have attempted to confirm this but have been unsuccessful. While the primary endpoints from these studies were not achieved, important conclusions can be derived from their results. In one trial, inhaled budesonide for 6 months in current smokers was unable to show regression of bronchial dysplastic lesions; however, a secondary endpoint, the proportion of CT-detected pulmonary nodules did show a slightly higher rate of resolution on imaging (87). This was followed by two subsequent phase II trials of inhaled corticosteroids that primarily examined resolution of CT-detected pulmonary nodules, yet could not identify a positive benefit (94, 95). However, there was a nonstatistically significant trend observed for resolution of nonsolid, and to a lesser extent, partially solid nodules. One reason hypothesized for the lack of an observed effect was that budesonide does not penetrate adequately into the peripheral lung (91). Analysis from six large lung screening trials suggest that while solid nodules carried a 7%–9% malignant potential, nonsolid nodules, which appear as ground-glass opacities on CT scan, were more likely to be malignant in 59%–73% of cases and were predominantly linked to the future development of bronchoalveolar carcinoma (96). It is difficult to ascertain the actual identity of ground-glass lesions on CT scans without subjecting patients to repeated bronchoscopies and invasive biopsies for histologic analysis, but this class of nodules likely represents atypical alveolar hyperplasia in most cases, the presumed precursor for pulmonary adenocarcinoma (97). In addition, these lesions have a long volume-doubling time and may defy the traditional rule that 2-year radiographic stability suggests a benign process (98). However, the lack of pathologic correlation with histologic biopsies remains a significant limitation in lung cancer prevention trials.
A different approach involves prostacyclin therapy. COX-2 is overexpressed in human lung carcinomas and in precursors leading to this malignancy (99). However, reducing prostaglandin synthesis by NSAIDs that selectively and nonselectively block both COX enzymes has yielded mixed results in animal models (100, 101) as well as in large human trials (73). Manipulation of the arachidonic acid pathway downstream from COX has become an attractive target for lung cancer prevention, as increased production of prostaglandin E2 has been observed to be carcinogenic and promote progression and metastasis through its direct effects on normal lung tissue and microenvironment (102). Increasing lung prostacyclin levels that counteract prostaglandin E2 through lung-specific prostacyclin synthase overexpression, downstream from cyclooxygenase, was found to be beneficial in some murine models, suggesting a conceivable mechanism of carcinogenesis and target for lung cancer prevention (103). Other research suggests that prostacyclin's protective effect is provided through a PPARγ-mediated mechanism (104).
In a phase II trial in humans with endobronchial dysplasia, oral iloprost, a stable synthetic prostacyclin analogue, was able to generate an improvement in bronchial epithelial premalignant dysplasia in former smokers, but not in current smokers (85). After the discontinuation of the oral form, the focus has shifted to delivering the drug through inhalation therapy. Inhaled iloprost is currently approved to treat pulmonary arterial hypertension with a 6–9 times per day dosing regimen that stimulates continuous vasodilation. This dosing schedule, however, would not be realistic for chemoprevention purposes. Our group therefore initiated a trial (NCT02237183) to investigate the utility of inhaled iloprost in preventing lung cancer in former smokers in a more practical twice per day and four times per day regimens.
Bexarotene, a retinoid that selectively activates retinoid-X receptors, is another promising preventive agent that has been explored in clinical trials against lung tumorigenesis. Unlike the previous beta-carotene and retinol studies that possessed minimal epidemiologic data, investigations into bexarotene were inspired by more substantial preclinical studies. Reduced expression of retinoid-X receptor-β (RXRβ) in some NSCLC biopsy specimens (105) and increased survival in patients with NSCLC (106) with increased RXRβ expression help illustrate the role of retinoid receptor activation in lung carcinogenesis. Despite promising cancer-inhibitory effects in both preclinical models (107) and early human trials (108), bexarotene administration induces significant hyperlipidemia with oral administration and was unable to demonstrate a survival benefit when combined with first-line therapy in chemotherapy-naïve patients with advanced or metastatic NSCLC (109, 110). In one preclinical model, aerosolized bexarotene demonstrated anti-proliferative and proapoptotic effects in vivo and significantly decreased tumor size and progression without increasing lipid levels (111). These results indicate that aerosolized bexarotene could provide a more effective means of retinoid delivery with reduced toxicity and may represent an attractive approach for future lung cancer chemoprevention trials.
Colorectal Cancer
Background: colorectal cancer
As the second leading cause of cancer-related deaths, and the third most common cause of cancer in the United States, colorectal cancer remains one of the deadliest cancers affecting the United States population today (74). There is approximately a 1 in 20 lifetime risk that a U.S. adult will develop colorectal cancer, with 90% of cases occurring in adults over 50 years of age (72). Accounting for 40% of new diagnoses, localized-stage disease carries a 5-year survival rate of close to 90% (112). Yet, with regional and distant metastasis, this number declines to 70% and 13%, respectively. Therefore, approximately one-half of those with newly diagnosed with colorectal cancer will die from it. Certain individuals are at much higher risk than the general population, including those with inflammatory bowel disease (113) and those belonging to families whose members are affected by familial adenomatous polyposis (FAP) and hereditary nonpolyposis colorectal cancer (HNPCC), in addition to those belonging to families with a cancer family syndrome, not otherwise specified (114).
Early-stage disease can present with subtle symptoms or only be first recognized during screening. However, the development of classic symptoms with anemia, hematochezia, weight loss, and abdominal pain signal a more advanced and less curable stage of disease. In an effort to reduce the number of advanced cases being first identified by the unexpected onset of symptoms, particular attention has been directed to the early detection of colorectal cancer in its premalignant stages.
The majority of colorectal cancer has been discovered to develop from the adenoma–carcinoma sequence, where over an average of 10 years, a polyp progresses from a small to large (>1 cm) polyp to later become a dysplastic, and finally a cancerous lesion (115). During this window period, effective preventative strategies have the potential of detecting and effectively reducing the risk of colorectal cancer. Following the National Polyp Study (NPS) that reported a 90% reduction in the incidence of colorectal cancer following polypectomy (116), a long-term follow-up study established a mortality benefit from the removal of adenomatous polyps (117). Currently, average-risk adults over the age of 50 years have several options for colorectal cancer screening based off of current screening guidelines (118). Colonoscopy is considered the gold standard and is particularly advantageous by providing a method for direct visualization and tissue sampling of suspicious lesions and polyp removal. Nevertheless, endoscopic evaluation is expensive and carries its own risks, including bowel perforation and serious bleeding, making this procedure not suitable to all candidates. Other less effective but proven alternatives include fecal analysis for occult blood and/or DNA, radiologic-based analysis (CT colonography and barium enema), and flexible sigmoidoscopy.
Despite the proven benefit of colorectal cancer screening, screening rates remain undesirably lower than anticipated with many factors being implicated including patient reluctance, discomfort, embarrassment, cost, and others (119). However, there has been an increase in screening rates over the past decade, where among the eligible adult population, rates have risen from 19% in 2000, to 55% in 2010 (120). As clinicians continue to encourage patient's involvement in colorectal cancer screening and efforts focus on overcoming the socioeconomic factors that impede complete participation, this percentage will likely increase.
While the screening has reduced colorectal cancer incidence, it does not completely remove risk, and other methods to further reduce incidence is the focus of ongoing investigations. In addition, research also implies the effectiveness of endoscopic evaluation may be limited in mitigating the risk of right-sided colorectal cancer (121). Colorectal cancer may also develop from other less noticeable nonpolypoid (flat) lesions (which have a right sided predominance; ref. 122), or, through the concept of field effect, develop from surrounding areas of mucosa in the peritumoral area that appear morphologically normal but have biologically altered epithelial cells that have already undergone multiple steps of genetic changes (123). These subtle lesions may not be detected by standard screening methods and remain elusive to our currently available screening methods (122).Chemoprevention represents an alternative approach to reducing the incidence and mortality risk of colorectal cancer that would complement preventative screening.
A product of our greater understanding of the molecular biology and genetic changes involved in carcinogenesis, chemoprevention provides the means of interrupting early stages of carcinogenesis. Many agents have shown to have great promise in experimental and case–control trials, but when tested in large prospective clinical trials they have been limited by either a lack of efficacy or by unacceptable levels of toxicity. As such, there are no currently available cost-effective therapeutic agents for colorectal cancer prevention for average-risk adults.
The most promising research involves the use of NSAIDs as preventative agents, which target the well-established role of inflammation in colorectal carcinogenesis, and act to inhibit COX-dependent pathways in driving inflammation and carcinogenesis. Initial epidemiologic studies suggested an inverse relationship between NSAID use and reduced colorectal cancer incidence and mortality. This generated significant interest, which led to further investigation with large population studies that reinforced this association. Systematic reviews combining controlled and observational studies evaluating the regular use of aspirin or NSAIDs came to the conclusion that their benefits were most apparent when used in high doses for periods longer than 10 years (4). Nonetheless, these large-scale studies identified significant risks with NSAID consumption, including increased rates of upper gastrointestinal complications with all NSAID regimens and increased cardiovascular events, particularly with selective COX-2 inhibitors, compared with placebo groups (124).
Annually in the United States, there are more than 100,000 NSAID-related hospitalizations and 16,500 deaths, which emphasizes that NSAID consumption is not benign (125). Efforts to reduce the adverse events associated with nonselective COX inhibitors led to further investigations using better tolerated COX-2–specific inhibitors, which have a more favorable side-effect profile. Similar to traditional NSAIDs, selective COX-2 inhibitors showed a consistent benefit when used as chemopreventive agents in patients with FAP (126, 127), HNPCC, and in patients with a recent history of adenoma resection (128). However, their benefit is eclipsed by the increased risk of cardiovascular events, thereby limiting their FDA indication only to patients with FAP who are inherently at high risk for colorectal cancer (129).The combination of sulindac–dimethyl sulfoxide (DFMO) treatment offers a promising alternative method by capturing the efficacy of NSAIDs as a chemoprotective strategy, but with lower levels of toxicity compared with other NSAIDs. One study reports a 70% reduction in recurrent adenomas, compared with placebo, with no significant increase in adverse events (130). The risk–benefit ratio in determining the clinical effectiveness of NSAIDs has been the focus of extensive debate. As of 2016, the United States Preventive Services Task Force (USPTF) now recommends low-dose aspirin for the primary prevention of cardiovascular disease and colorectal cancer in adults ages 50–59 years who have a 10% or greater 10-year cardiovascular disease risk, are not at increased risk for bleeding, have at least a 10-year life expectancy, and are willing to take low-dose aspirin daily for at least 10 years (131). Outside of this setting, their use as a preventative agent remains limited to their consideration in those at highest risk for colorectal cancer. Their use is not recommended for the general population at this time.
In a series of studies evaluating the benefit of hormone replacement therapy (HRT) in chronic disease prevention in women, including results from the landmark comprehensive Women's Health Initiative, a reduced risk of colorectal cancer was observed in women who received HRT (132, 133). However, further analysis highlighted a greater detrimental effect with increased risk of cardiovascular death and increased breast cancer rates, as well as a paradoxically increased colorectal cancer–related mortality risk despite the lower incidence (134). The increased colorectal cancer–related fatality rate was attributed to a finding that while there was a reduced incidence of colorectal cancer with combined hormonal therapy, the rate of metastatic cancer was found to be higher in the treatment group, suggesting a diagnostic delay in detecting advanced disease.
Other agents have been studied in a prevention setting, but have failed due to lack of efficacy. Initial epidemiologic studies and preclinical models that suggested a chemopreventive potential with supplemental Vitamin D and calcium were found not to significantly reduce the risk of colorectal cancer when used in combination (135) or separately (136). Similarly, studies examining a protective effect from folic acid (137), fiber (138) and antioxidants (including, vitamins A, C, E, beta-carotene, and selenium; ref. 139), also failed to demonstrate a positive benefit.
There are a number of challenges with analyzing the clinical effectiveness of the aforementioned interventions in the preventative research setting. Given the insidious onset of colorectal cancer by its long premalignant phase as it develops from an adenoma to carcinoma, studies experimenting with specific agents or interventions and their effect on colorectal cancer incidence and mortality can be difficult to assess outside of studies with long-term follow-up. Therefore, it is important to note that the trials looking into the utility of chemopreventive agents have primarily looked at surrogate endpoints, such as the incidence of recurrent adenomas or the development of aberrant crypt foci (ACF), a presumptive precursor lesion to adenoma and cancer.
The use of recurrent adenomas as a preventive endpoint is supported by retrospective and prospective studies demonstrating that they serve as practical markers for colorectal cancer incidence and mortality (140). However, the limitations to using adenomas as biomarkers include the relative infrequency of transformation of small adenomas to cancers, cost-effectiveness of an endoscopic approach for cancer prevention, and the fact that some cancers arise from flat or depressed lesions that would be otherwise missed in studies that set recurrent polyps as endpoints (141). Many preclinical and current clinical trials have also drawn attention to ACF and their link to future cancer, particularly those that are large and dysplastic (142). Increased proliferative activity and K-RAS mutations in ACF draw an even closer connection between ACF and early colon cancer (143). However, despite strong evidence to support the use of alternate endpoints, the significant correlations demonstrated between biomarkers and their future oncogenic potential is not proof of a direct link to cancer, and their use is still debated.
Approach to prevention: local delivery agents
The concept of local prevention can be extended to the gastrointestinal tract. Therapeutic agents that are poorly absorbed yet locally alter colonic dysplastic cells from progression to neoplasia would offer an attractive method of delivering a potent antineoplastic agent that remains in the gastrointestinal tract without prompting the sequelae of systemic absorption and toxicity. Also, agents that are absorbed, but undergo a high degree of enterohepatic circulation provide for a situation wherein pharmacologic exposure is regionally localized. Under such circumstances, systemic absorption is low to nil, as is systemic toxicity, while the target organ is exposed to high concentrations of the therapeutic agent.
Polyethylene glycol (PEG) is a nonabsorbable, nonmetabolized, water-soluble polymer currently used as lavage solutions for bowel cleansing prior to colonoscopies and in those with occasional constipation. It is currently being looked at with an innovative approach as a tool for colorectal cancer prevention research. Research into PEG as a chemopreventive agent has been performed over several decades. Initial preclinical studies in rats led to the discovery that PEG induced a dose-dependent suppression of ACF (144) and that it functions as one of the most potent agents for inhibiting carcinogenesis in a prevention setting (145), outperforming NSAIDs, fiber, vitamin D, and ursodeoxycholic acid (146). PEG can deliver a fast, consistent, and potent inhibitory effect on early colonic precursor lesions within several days of treatment; however, the suppressive effect was reversible with drug discontinuation (147). Among other laxatives, PEG was the only one that provided a benefit on reducing ACF and for inducing apoptosis of abnormal mucosal cells, suggesting its benefit was not directly correlated with its laxative ability (148).
The combination of PEG's promising benefits in preclinical models for preventing colorectal cancer and excellent side-effect profile with three decades of use has helped foster significant interest into its use as an antineoplastic agent. Its efficacy, at standard laxative dosing (17 g of PEG 3350) induces a modest increase in daily bowel movements by less than a half bowel movement per day over that observed with placebo, with no reported difference in adverse events between study and placebo groups (149). While side effects including nausea, abdominal bloating, cramping, diarrhea, and flatulence may occur, short-term (149) and long-term (150) studies have shown that it is a safe and effective drug with no directly related serious effects. Unlike the smaller molecular weight forms, PEG 3350 has a low risk of systemic toxicity as a result of virtually no absorption (<0.28%) and rapid elimination via feces (151).
The utility of PEG in human colorectal cancer prevention has yet to be determined. In a French population-based study, observing patients attending for routine total colonoscopy, patients who had used Forlax (PEG 4000) at any time in the past were found to have half the risk of developing colorectal tumors suggesting that PEG's efficacy in chemoprevention may be extended to humans (152). To capitalize on the series of promising preclinical and epidemiologic data, our group has implemented and recently completed a clinical trial evaluating standard-dose PEG in colorectal cancer prevention (NCT00828984), with resultant data currently undergoing analysis.
Similar interest has been invested into the use of berberine in colorectal cancer prevention in patients with ulcerative colitis. Individuals with ulcerative colitis suffer from chronic colonic inflammation that places them at a higher risk of colorectal cancer with an incidence that is 2- to 5-fold higher than that of the general population, with risk increasing throughout life after diagnosis of ulcerative colitis (2% at 10 years, 8% by 20 years, and 18% by 30 years; ref. 153). Longstanding and extensive colitis (LEC), defined as colitis for >10 years and involving >50% of the colon, was also associated with an increased risk of colorectal cancer as compared with the general population, which emphasizes the significance of controlling inflammation in the disease process (154). Furthermore, once colorectal cancer develops, treatment is not as effective in those with ulcerative colitis, as compared with the general population, and is associated with a poorer prognosis in patients with ulcerative colitis than those without (155).
An approach that targets the proinflammatory pathways can be seen as a central tenet to colorectal cancer prevention in patients with ulcerative colitis. Berberine, the active constituent of the Coptidis rhizoma plant, has been commonly used in traditional Chinese medicine for thousands of years to treat gastrointestinal disorders, as well as inflammatory disorders, including arthritis (156). Recently, it has been discovered to inhibit inflammation-triggered carcinogenesis (157) and while the exact anticancer mechanisms by which berberine is able to effectively inhibit cancer cell proliferation are not fully understood, effects upon inflammation appear important (158). Berberine promotes recovery of colitis and exerts inhibitory effects on the proinflammatory responses in colonic macrophages and epithelial cells. Berberine also effectively modulates several biomarkers relevant to cancer, including those related to proliferation (cyclin D1), inflammation (COX-2), invasion (matrix metalloproteinase-9), and anti-apoptosis (Bcl-xL, Survivin, IAP1, IAP2, and cFLIP; ref. 159). In animal models, berberine was found to prevent neoplastic transformation, in association with induction of the antioxidant defense system and its activation of proapoptotic properties, and effectively prevented tumor occurrence (160). In addition, in vitro studies using human colon cancer cells found that berberine could significantly suppress tumor growth in a dose- and time-dependent manner (161).
In trials assessing its safety in treating acute watery diarrhea, berberine was found to be effective alone as well as in combination with antibiotics (162). Currently, it is approved by the Chinese Food and Drug Administration (CFDA) and is available as an over-the-counter treatment for acute gastrointestinal ailments. In murine-based studies, these conventional doses were associated with an extremely low oral bioavailability of approximately 0%, mainly due to rapid first-pass metabolism and excretion by the liver (163, 164). In multiple clinical trials, including adults in the Asian population, berberine at a dose of 1.0–1.5 g daily for 3 months has been found to be safe, and the herbal derivative is not considered toxic at these doses (165, 166). Because genetic factors are major determinants of drug metabolism, extrapolating pharmacokinetics and safety information across populations is complex. Therefore, berberine is only available in the United States as a dietary supplement and is not approved by the FDA.
The combination of berberine's anti-inflammatory and antineoplastic features in preclinical investigations, preclinical studies demonstrating enterohepatic circulation and low levels of systemic delivery, the pathologic role of chronic inflammation in the development of colorectal cancer in ulcerative colitis patients and an extensive safety record in the Chinese population, coupled to its unique and favorable pharmacologic profile, which involves relatively selective delivery of drug to the target organ, has garnered interest in the cancer prevention research community. This has led our group to implement a prospective trial (NCT02365480) exploring the safety and efficacy on tissue-based measures of inflammation and dysplasia in Chinese ulcerative colitis patients who are in clinical remission while receiving maintenance mesalamine.
Head and Neck Cancer
Background: head and neck cancer
Most head and neck cancers begin from the mucosal surface of the upper aerodigestive tract and are predominantly squamous cell by histology. Chronic oral exposure to tobacco (smoked and smokeless), alcohol, and/or HPV results in hyperplasia of squamous epithelium, which may transform to oral dysplastic and precancerous lesions known as oral erythro- or leukoplakia, which present as red or white plaques involving the oral mucosa. While oral erythro- or leukoplakia is benign, there is up to a 20% risk of progression to cancer (167). Given the strong association between smoking and alcohol and the development of head and neck cancers, cessation of alcohol and tobacco products and maintaining abstinence must also be incorporated into the approach to treatment.
EGFR is highly expressed in these oral premalignant lesions and is a strong predictor of progression to cancer, suggesting that receptor blockade may be effective in preventing progression to invasive cancer (168). The concept of field cancerization, whereby patients with head and neck squamous cell carcinoma (HNSCC) lesions have a tendency for cancer development throughout the oropharyngeal tract, even in normal appearing mucosa, provides a strong rationale for chemoprevention (169). Inhibition of the EGFR signaling pathway, however, was found to be ineffective in a phase III clinical trial evaluating the use of the EGFR inhibitor, erlotinib, in patients with oral dysplastic lesions and failed to confer a reduced cancer incidence compared with placebo (170).
Approach to prevention: Local delivery strategies
Treatment with photodynamic therapy (PDT) with the photosensitizing agent aminolevulinic acid (ALA) has also been studied by our group in a phase I clinical trial for the treatment of oral leukoplakia which primarily looked at safety and tolerability with secondary endpoints examining clinical response to treatment (171). Orally administered ALA followed by light stimulation (which can be locally directed, in this case to the oral lesion), using laser light corresponding to an optimal wavelength (410–635 nm), photoactivates ALA, and thereby produces reactive oxygen species capable of inducing local cytotoxic damage. ALA has several advantages that make it a suitable agent in chemoprevention. It is an endogenous metabolite present in all mammalian cells involved in the early steps of heme synthesis and is a naturally occurring precursor to the endogenously formed photosensitizer, protoporphyrin IX (172). In addition, rapid excretion (over 24–36 hours), selective concentration in mucosal and epithelial tissues rather than deeper stromal tissue (173), and possibly preferential accumulation in dysplastic and tumor cells rather than normal tissue (174) all serve to provide a pharmacologic profile wherein drug is selectively accumulated in the target tissue, and relatively rapidly excreted from the body. Although, insufficient patient accrual led to early termination of the study, the study yielded important findings. Of 11 evaluable patients treated, who were receiving escalating amounts of light intensity (i.e., “doses of light”) in the context of a phase I dose escalation trial, resolution of a lesion was observed in one patient. Toxicity was transient and minor. Together, these findings support the notion that ALA-PDT therapy for the treatment of oral leukoplakia holds promise and should be investigated further. Furthermore, while ALA-PDT could be effective in targeting oral premalignant lesions, this therapy does not eliminate the possibility of additional cancer development attributed to field cancerization. This was demonstrated by a study by Lee and colleagues, who observed a cohort of patients with oral leukoplakia, where 59% of patients developed cancers at the same site as prior cancer or leukoplakia while 41% developed cancers away from prior lesions (167). This would suggest that local ablative-type of therapy will need to be repeated in the context of ongoing surveillance, and/or may need to be combined with systemic therapy to achieve the best cancer prevention effect.
Recognizing that EGFR has been implicated in colon carcinogenesis as well as in head and neck carcinogenesis, and that PEG is a potent downregulator of EGFR signaling, one group has gone on to demonstrate remarkable results in murine models with PEG administration (175). These findings will support future investigations in humans at risk for head and neck cancer.
Esophageal Cancer
Background: esophageal cancer
The majority of esophageal cancers belong to two predominant histologic subtypes: esophageal squamous cell cancer (ESCC) and esophageal adenocarcinoma. The incidence of each varies among geographic regions. ESCC is most prevalent in developing countries, while the incidence of esophageal adenocarcinoma arising out of Barrett esophagus has risen dramatically in Western nations (176). Barrett esophagus develops in response to chronic exposure to gastric acids as a complication of longstanding gastroesophageal reflux disease (GERD) resulting in the replacement of the normal squamous esophageal lining with intestinal epithelium, and this predisposes to esophageal adenocarcinoma. The recent epidemic of obesity present in the United States, and other Western nations, predisposes to increased abdominal pressure, resultant GERD, and may thereby ultimately represent a primary cause of the recently observed increase in esophageal adenocarcinoma in Western nations. The risk of cancer from Barrett esophagus is poorly defined but is thought to be determined by the degree of dysplasia with high-grade dysplastic lesions having close to 6% risk per year to progression to esophageal adenocarcinoma (177).
In a meta-analysis that included 2,813 patients with Barrett esophagus and 317 patients with high-grade dysplasia or esophageal adenocarcinoma, the use of proton pump inhibitors (PPI) resulted in a reduced risk of high-grade dysplasia and/or esophageal adenocarcinoma (adjusted OR 0.29; 95% CI, 0.12–0.79). Therefore, all patients with Barrett esophagus are recommended to receive treatment with a PPI. In addition, secondary prevention of esophageal adenocarcinoma consists of endoscopic screening for the purposes of early detection and subsequent treatment of precancerous lesions or early-stage cancer. Any mucosal irregularities seen on initial endoscopy are biopsied and removed by endoscopic mucosal resection (EMR). Subsequent therapy is determined on the basis of the degree of dysplasia seen on histopathologic examination.
Surveillance every 3 to 5 years is generally recommended for those with nondysplastic Barrett esophagus. However, any diagnosis of dysplasia or intramucosal carcinoma is usually managed with endoscopic ablative therapy, including the use of thermal, photochemical, or radiofrequency energy to ablate abnormal mucosal tissue in Barrett esophagus. Today, the preferred method for endoscopic eradication therapy consists of EMR of any visible mucosal irregularities, followed by radiofrequency ablation (RFA) of the remaining metaplastic epithelium, provided that there is no evidence of submucosal invasion which would warrant more aggressive therapy. This approach is the most effective and favored of endoscopic ablative procedures due to a better safety profile and superior outcomes in reducing the risk of neoplastic progression and resulting in eradication of both low-grade (178) and high-grade dysplastic lesions in 92% of cases (179). However, one study of 246 patients with high-grade dysplasia or intramucosal carcinoma with initial complete eradication of all intestinal metaplasia in 80% of cases, recurrence of neoplasia was seen in 25% at 60 months and recurrence of metaplasia was seen in 50% at 48 months (180). This serves to highlight the importance of continued endoscopic surveillance despite initial favorable eradication rates.
Approach to prevention: local therapeutic strategies
Photodynamic therapy (PDT) with the photosensitizer, porfimer sodium (PS, Photofrin), was first approved by the FDA in 2003 for treatment of high-grade dysplasia in patients with Barrett esophagus. Its approval was based upon the results of a randomized controlled trial of 208 patients with high-grade dysplasia that were randomized to receive PDT PS with omeprazole or omeprazole alone (181). A statistically significant improvement was observed with complete ablation of high-grade dysplasia in 77% in the treatment arm versus 39% in the placebo arm. Treatment-related complications, however, including serious complications (12% vs. 1%) were higher in the PDT PS arm; 36% of patients in the PDT PS arm developed strictures, although the majority (98%) resolved with endoscopic dilation.
PDT ALA has also been studied to treat Barrett esophagus. One study of PDT ALA in 32 patients with high-grade dysplasia or superficial mucosal cancer resulted in complete eradication of high-grade dysplasia and tumors less than 2 mm thickness in all patients with no procedure–related morbidity or mortality (182). PDT ALA was also examined in patients with low-grade dysplasia. In a double-blind, randomized controlled trial of 36 patients, patients were assigned to receive ALA or placebo followed by laser endoscopy. All patients who received ALA had complete clearance of dysplasia and 89% showed macroscopic evidence of Barrett esophagus regression compared with 33% clearance of dysplasia and 11% reduction in the area of Barrett mucosa among those in the placebo arm (183).
In comparison with PS, ALA has several advantages including oral administration and shorter duration of skin photosensitivity (24–36 hours vs. 30 days; ref. 173). In addition, ALA exhibits preferential mucosal accumulation, potentially leading to fewer complications due to shallower tissue injury compared with PS, which accumulates in the submucosa. However, the deeper accumulation of PS in the submucosa may be required to penetrate the full thickness of the dysplastic tissue.
Bladder Cancer
Background: bladder cancer
After prostate cancer, urothelial (transitional cell) carcinoma of the bladder is the second most common malignancy of the genitourinary tract and predominantly arises from transitional cells that line the urinary tract. Approximately 70% of bladder cancer lesions are classified as non-muscle–invasive bladder cancer (NMIBC), which may be viewed as precancerous, similar to other organs (184).
At the time of diagnosis, transurethral resection of all visible bladder tumor (TURBT) using electrocautery is the initial treatment of NMIBC and is used to determine clinical staging, which sequentially drives subsequent treatment strategy. After diagnosis of NMIBC is confirmed, patients are risk stratified from low to high risk to identify those patients that would benefit from more aggressive therapy, including cystectomy. Important factors including staging, histologic grade, presence of multicentric disease, and presence of carcinoma in situ portend higher risk of disease, and serve to guide risk stratification and clinical management. In patients with NMIBC, 40%–80% will have disease recurrence within 12 months when treated exclusively with TURBT, and 10%–25% of patients will develop muscle-invasive, regional, or metastatic disease (185–187). As such, more aggressive therapy is usually pursued after TURBT. Therefore, for patients discovered with NMIBC at the time of initial cystoscopic examination, intravesical chemotherapy or immunotherapy is typically administered in conjunction with TURBT to delay or prevent disease recurrence or progression to muscle-invasive disease.
Approach to prevention: Local therapeutic strategies
Intravesical therapy allows for the instillation of high local concentrations of a drug within the bladder with the purpose of destroying viable tumor cells remaining after TURBT and preventing tumor implantation. The Bacillus Calmette-Guerin (BCG) vaccine and valrubicin are two agents approved for the treatment and prophylaxis of carcinoma in situ (Tis) of the bladder and for the prophylaxis of primary or recurrent stage Ta and/or T1 papillary tumors after TURBT. BCG, which was initially developed as a vaccination against tuberculosis, activates a variety of local immune responses after intravesical instillation, which appear to correlate to antitumor activity (188, 189). While a number of other intravesical agents have been compared with BCG in head-to-head studies, none have been shown to be as effective, and BCG remains the standard of care following TURBT for high-risk NMIBC (190) as it has been shown to reduce tumor recurrence and to improve overall survival (191, 192).
Single-dose intravesical chemotherapy, most commonly with mitomycin C, in the immediate postoperative period for low-risk papillary disease, significantly decreases the risk of recurrence after TURBT in patients with stage Ta, T1 single and multiple bladder cancer (193). In patients with low-risk disease, TURBT plus a single dose of intravesical therapy during the initial procedure is usually sufficient. Those deemed to have an intermediate or high risk of recurrence benefit from receiving multiple intravesical chemo- or immunotherapy treatments, and subsequently may be considered for a restaging TURBT. This is supported by one systematic review that compared six trials assessing the utility of intravesical BCG with TURBT compared with TURBT alone in patients with Ta and T1 bladder cancer of intermediate- or high-risk for tumor recurrence. This study showed lower rates of disease recurrence in 26% of patients at 12 months in the BCG plus TURBT group versus 51% in the TURBT alone group (194).
BCG can induce complete response rates as high as 80% in patients with high-risk NMIBC. However, an estimated 50%–70% of high-grade T1 lesions and up to 90% of Tis tumors recur by 5 years (195). The extent of posttreatment follow-up and management that follows intravesical therapy can be complex and would be beyond the scope of this article. Nevertheless, cystectomy is indicated in those discovered to have muscle invasion (T2 or greater) and in select cases of NMIBC as a prophylactic measure, including those with recurrent or persistent disease within 6 to 12 months of intravesical therapy, patients determined to be at high risk for progression to muscle-invasive disease including lesions with evidence of carcinoma in situ and/or deep involvement of the lamina propria and for palliation in some patients experiencing treatment-related complications such as intolerable urinary frequency, pain, incontinence, and hemorrhage. The anthracycline valrubicin is also approved in the United States for intravesical use in the second line of treatment for patients with Tis lesions who have persistent or recurrent non-muscle–invasive disease refractory to intravesical BCG and in whom immediate cystectomy would be associated with unacceptable morbidity or mortality.
Summary
Over the last century, our approach to cancer has transformed tremendously and will likely continue to transition from focusing on treating invasive cancer to having a greater emphasis on early detection and prevention. Local delivery strategies for chemoprevention have already been shown to be highly effective in several organ types. In addition to efficacy, and of very high importance in the context of a prevention strategy, is the fact that local delivery strategies selectively avoid systemic delivery and the associated toxicity. Recent advances have demonstrated the feasibility of local delivery strategies in organ types heretofore not considered, and have done so in organs that constitute major sites of cancer development, that in turn have major public health implications. Coincident with the development of local therapeutic strategies is the rapid development of technologies increasingly able to identify and refine high-risk cohorts, and to assess in real time their response to therapeutic interventions.
Cancer chemoprevention based upon a local therapeutic delivery strategy offers an extremely high impact potential. That potential will clearly be enhanced through future developments in nanotechnology and associated delivery platforms. However, existing technology is clearly capable of moving the field forward at this time. A central element in realizing the true potential of this strategy relates to rationally building upon current successes. If brought forward in a well thought out rigorous manner, the lives of many people could be positively impacted in the future.
Disclosure of Potential Conflicts of Interest
No potential conflicts of interest were disclosed.
Disclaimer
The authors listed above certify that they have no affiliations with or involvement in any organization or entity with any financial interest (such as honoraria; educational grants; participation in speakers' bureaus; membership, employment, consultancies, stock ownership, or other equity interest; and expert testimony or patent-licensing arrangements), or non-financial interest (such as personal or professional relationships, affiliations, knowledge or beliefs) in the subject matter or materials discussed in this article.