Phase I and Pharmacology Study of Ropidoxuridine (IPdR) as Prodrug for Iododeoxyuridine-Mediated Tumor Radiosensitization in Advanced GI Cancer Undergoing Radiation

Radiation therapy (RT) is a critical component of cancer therapy, in both the curative and palliative settings, with approximately 60% of patients receiving RT at some time during the course of their cancer treatment (1). The availability of additional radiosensitizing drugs that improve the efficacy/toxicity profile of RT would represent a meaningful advance in cancer therapy, potentially improving survival and lessening morbidity for patients with cancer in a wide range of clinical situations. Indeed, the development of radiosensitizers has been targeted as a priority in a collaborative project involving the NCI, the Radiation Therapy Oncology Group (RTOG; now part of NRG Oncology Group) and investigators from the United Kingdom (2, 3).

A number of systemic agents are used in clinical practice to enhance the effect of RT (2), including platinum compounds (4–6), signal transduction inhibitors (7), fluoropyrimidines (8–11), gemcitabine (12), and temozolomide (13). The use of cytotoxic agents with RT has been found to benefit patients with select tumor types, for example temozolomide in glioma, fluoropyrimidines in gastrointestinal (GI) malignancies, and cetuximab/EGFR for head and neck cancers. Although attempts have been made to exploit tumor hypoxia to improve radiation response, no agents in this class are currently considered standard-of-care (14). Broader use of these agents has been limited by toxicity, especially in the context of multimodality, multiagent treatment regimens. These existing examples validate the clinical proof-of-principle of radiosensitization and encourage development of additional radiosensitizing agents with higher therapeutic indexes.

5-Iodo-2-pyrimidinone-2′-deoxyribose (IPdR) is a prodrug of iododeoxyuridine (IUdR), a pyrimidine analog that has been recognized as a radiosensitizing agent since the early 1960s (15, 16). IPdR offers several key advantages compared with IUdR, including a more favorable therapeutic index.

The mechanism of IUdR-mediated radiosensitization involves the cellular uptake and metabolism of IUdR through the thymidine salvage pathway. IUdR undergoes initial intracellular phosphorylation to the monophosphate derivative by the rate-limiting enzyme, thymidine kinase, followed by sequential phosphorylation to the triphosphate. This modified analog is then utilized (i.e., incorporated) by DNA polymerase during scheduled (S-phase) and unscheduled DNA synthesis, in competition with thymidine triphosphate. DNA incorporation is needed for radiosensitization by ionizing radiation (IR), in human tumors as well as normal cells, and the extent of radiosensitization correlates directly with the %IUdR DNA replacement (17–19). IR-mediated radiosensitization by IUdR results in the generation of highly reactive uracil free radicals by IR, which may also damage unsubstituted complementary-strand DNA, resulting in increased DNA single strand breaks and increased double-strand breaks (18–20). It is also known that the repair of IR damage may also be reduced by pre-IR exposure to IUdR. IUdR is rapidly metabolized in both rodents and humans when given as a bolus infusion with a plasma half-life of <5 minutes (15). Consequently, prolonged continuous drug infusion over several weeks before and during RT is necessary to maximize the proportion of tumor cells that incorporate IUdR during the S phase of the cell cycle (21, 22).

In the 1980 to 1990s, clinical trials of IUdR as a radiosensitizer were performed, focusing on patients presenting with high-grade gliomas (23–26) and sarcomas (27, 28). These studies demonstrated that the magnitude of radiosensitization correlates directly with plasma IUdR levels and the %IUdR-DNA tumor cellular replacement, and that both measurements can serve as radiosensitization biomarkers (29, 30). Additionally, in small series of patients with head and neck cancers or liver metastases from colorectal cancer, the %IUdR-DNA cellular incorporation in tumors ranged to 5%, but was less than 1% in adjacent normal liver tissue, further supporting a therapeutic window for IUdR-mediated radiosensitization (30–32). Although IUdR has clear activity as a radiosensitizer, it requires prolonged continuous infusion before and throughout the course of RT, and systemic toxicities (myelosuppression, GI toxicities) limit the tolerable dose and thus IUdR's potential for radiosensitization. These issues have constrained the development of IUdR as a radiosensitizer (23–26).

IPdR is a prodrug of IUdR, which is administered orally and then efficiently converted to IUdR by a hepatic aldehyde oxidase (33). Although preclinical studies have many limitations, the broad range of pharmacology and safety testing in mice, rats, ferrets, and Rhesus monkeys has been promising in terms of the ability of IPdR to mediate radiosensitization with acceptable toxicity (34–38). Strong support for advancement of IPdR to the clinic was provided by radiosensitization studies in vivo using human tumor xenograft models: 2 colorectal cancer models (HT29, HCT-116), and 2 glioblastoma (GBM) cell lines (U251 and U87; refs. 34, 35, 37, 39, 40).

This collection of preclinical studies demonstrated that: (i) IPdR could be administered and tolerated orally (gavage) using multiple daily dosing up to 28 days; (ii) IPdR is efficiently and predictably metabolized into the active metabolite, IUdR; (iii) relatively prolonged plasma levels of IUdR can be achieved; (iv) %IUdR-DNA incorporation was demonstrated to be 2 to 3 times greater in proliferating tumor and 2- to 3-fold less in proliferating normal tissues (bone marrow and GI epithelium) compared with continuous infusion IUdR at the MTD; (v) radiation sensitizer enhancement ratios of 1.3 to 6 were achieved using human cancer xenograft models, and compared favorably with those of continuous infusion IUdR at the MTD (≤1.1); and (vi) systemic toxicity was manageable at effective doses. These data were used to support an Investigational New Drug Application (IND; #70,333) and to conduct an initial phase 0 trial of single-dose IPdR in patients with advanced malignancies that demonstrated radiosensitizing plasma levels of IUdR could be achieved following oral administration of IPdR (41).

The choice of patient population for this phase I and pharmacology trial was based on an NCI-sponsored IPdR Experts Conference that prioritized GI cancers and the NCI Colorectal Cancer Working Group's interest in combining IPdR with standard-of-care therapy for patients with rectal cancer (11). Patients undergoing intermediate-dose palliative RT for metastatic GI cancers received a single every day oral dose of IPdR beginning 7 days before the initiation of RT through the final day of RT. Plasma levels of IPdR, IUdR, and metabolites were measured at multiple time points throughout the 28 days, patients were monitored for toxicity, and tumor response was assessed at Week 8. The maximum tolerated dose (MTD) and the recommended phase II dose (RP2D) were established in this patient population. Results of this trial support phase II testing of oral IPdR as a radiosensitizing agent.

Patients 18 years and older with histologically or cytologically confirmed advanced, incurable cancers of the esophagus, liver, stomach, small bowel, pancreas, bile duct, colon, or rectum referred for palliative RT to the chest, abdomen, and/or pelvis were eligible for the trial. Patients received no other systemic therapy for 1 month prior to study and for the duration of study (8 weeks). Patients receiving palliative RT for advanced primary tumors or metastatic disease were eligible. Eligible patients had acceptable organ function based on laboratory data and an Eastern Cooperative Oncology Group performance status (ECOG PS) of ≤2. Patients must not have received prior RT to the tumor site being irradiated on this study. Exclusion criteria included brain metastases, a history of allergic reactions to compounds of similar chemical or biologic composition to IPdR, uncontrolled intercurrent illness, psychiatric illness/social situation that would limit compliance with study requirements and pregnancy or breast feeding. The trial was conducted under a NCI-sponsored IND (#70,333) and conducted by the Brown University Oncology Group (BrUOG) at Lifespan Rhode Island Hospital (RIH). The protocol and informed consent documents were approved by the Cancer Therapy Evaluation Program (CTEP) and the Lifespan-RIH IRB. Patients provided written informed consent prior to enrollment. Protocol design and conduct followed all applicable regulations, guidances, and local policies. The study is registered on ClinicalTrials.gov, NCT02381561.

This was a single-institution, single-arm dose escalation study. IPdR was supplied by the Developmental Therapeutics Program (DTP), NCI. Patients received a single every day oral dose of IPdR, beginning 7 days prior to the initiation of RT and continuing QD through the final day of RT (Days 1–7 without RT; Days 8–28 with RT; Fig. 1). IPdR was supplied as 75 and 300 mg capsules. Capsules were taken on an empty stomach (i.e., NPO for 2 hours prior to and 1 hour following oral administration). On the days of RT, IPdR was administered within 0.5 to 2 hours prior to RT. Patients kept a medication log, recording the date, time and number of capsules of each dose.

Patients received intensity modulated radiation therapy (IMRT; 4–18 MeV photons) to a total dose of 37.5 Gy using 15 2.5 Gy fractions delivered once daily on a Monday to Friday schedule during Weeks 2 to 4 of IPdR treatment. This intermediate-dose RT schedule is not typical for the palliative setting, where more commonly, a total dose of 30 Gy is delivered in 10 3 Gy fractions or 20 to 25 Gy is administered in 4 to 5 Gy fractions, and where 3-D planning is often the norm. The choice of RT dose, schedule and modality used in this trial reflected the goals of appropriate patient care and the generation of results that could be applied to the clinical setting of the use of RT for GI cancers. IMRT was used to limit the dose to adjacent normal tissues and to best quantitate any local (within the 50% isodose volume) toxicities of IMRT + IPdR using total radiation doses equivalent to those used in the pre- and/or postoperative settings for GI cancers (45–50.4 Gy over 5 to 6 weeks). The IMRT treatment plan for each patient was based on an analysis of the volumetric dose, including dose volume histogram (DVH) analyses of the planning tumor volume (PTV), defined as the clinical tumor volume (CTV) with a 7 mm margin. Ninety percent of the PTV received 95% of the prescribed dose, and 99% of the CTV received 95% of the prescribed dose. The gross tumor volume (GTV) was defined as the symptomatic metastatic or locally recurrent tumor based on contrast enhanced CT or MRI scans, and was expanded by 1.5 cm to define the CTV. The Rhode Island Hospital Department of Radiation Oncology is credentialed for IMRT use in the chest, abdomen, and pelvis through Imaging and Radiation Oncology Core (IROC) Houston for all Eastern Cooperative Oncology Group-American College of Radiology Imaging Network (ECOG-ACRIN), Radiation Therapy Oncology Group (RTOG), and Gynecologic Oncology Group (GOG) protocols (now NRG Oncology).

A primary objective of the study was to determine the safety, toxicities, and MTD of IPdR administered with concurrent RT. Clinical toxicities were graded according to revised NCI Common Terminology Criteria for Adverse Events (CTCAE) version 4.0 (version 5.0 when it became available). The study was designed with a 2-part dose escalation scheme. The starting dose was 150 mg every day x 28 days (≅1/10th MTD in ferrets; ref. 38). Part I was a single patient dose escalation of PO IPdR every day x 28 days, with 100% dose escalation for each sequential patient until a treatment related grade 2 systemic toxicity was observed. Three additional patients were then entered at this dose level. If a second patient of these 3 patients experienced a treatment-related grade 2 systemic toxicity (or if a single patient developed treatment related ≥grade 3 toxicity), then a stop/switch rule was invoked for part I, and part II of the dose escalation scheme commenced. If not, the dose was incremented for the next patient. In part II, 3 patients were treated at the final part I dose level, with a 40% dose increment schedule for subsequent dose levels. If 0 of 3 patients did not experience a ≥grade 3 treatment-related dose-limiting toxicity (DLT), accrual continued with 3 patients per dose level, until a ≥grade 3 treatment-related DLT was observed. DLT was reached when any 2 grade 3, systemic or local (within the 50% isodose volume of the radiation plan), treatment-related toxicities were observed in 2 of the up to 6 patients enrolled at that dose level. One additional patient was treated at the dose level below the dose associated with DLT to establish the MTD. The MTD was defined as the dose below which 2 or more of up to 6 patients experience DLT.

For the pharmacokinetic studies, 5 mL of blood was collected at 0, 30, 60, 120, and 240 minutes following the oral dose of IPdR on Days 1, 15, and 22, and a Day 1 24-hour sample was obtained immediately prior to the Day 2 dose. Following collection of blood, plasma was separated via centrifugation, and then shipped per protocol to the NCI Division of Cancer Treatment and Diagnosis (DCTD) Pharmacokinetics Laboratory at the Frederick National Laboratory for Cancer Research in Maryland. Procedures for pharmacokinetic analyses are available in Supplementary Materials and Methods.

In addition to internal review, external review of the raw data was conducted by Theradex and RTI Health Solutions in the context of generating the formal Clinical Study Report, which was reviewed by NCI-CTEP physicians.

Nineteen patients were enrolled on the trial. Patient demographics and characteristics are presented in Table 1.

Table 2 describes each patient's treatment, toxicity, and response.

In part I of the dose escalation scheme, the first and second patients received 150 and 300 mg orally every day, respectively, without DLT. The third study patient (600 mg orally every day) developed grade 3 transaminase elevation on Day 22 of study, in the context of presenting with fever and staphylococcus cellulitis at a previously removed mediport site (culture positive: staphylococcus). CT on Day 23 revealed an increase in the size and number of hepatic metastases (nontarget) and the size of the target lesion in the tail of the pancreas presumed to be progressive disease (PD), and the patient was removed from study on Day 24. At follow-up on Day 38, the cellulitis had resolved, transaminases were within normal limits, and no toxicities were noted. This event triggered the Stop/Switch rule, and part II of the dose escalation scheme was initiated.

Cohorts of 3 evaluable patients each for dose levels 600, 900, and 1,200 mg received therapy without evidence of IPdR-related grade 2 or higher clinical or laboratory toxicity. Four patients did not complete study therapy. One patient (1,200 mg) was removed from study on Day 2 per patient request. One patient (600 mg) developed PD after 22 days of IPdR and 12 days of RT, and discontinued all therapy. Two patients were removed from study before completion of IPdR therapy secondary to unrelated complications. One patient (600 mg) developed an incarceration of his colostomy on Day 9 (outside of RT field), and one patient (900 mg) developed a bowel obstruction from progressive colon cancer on Day 12 (outside of RT field). Both patients required emergent surgical intervention, and subsequently recovered. All patients were evaluable for toxicity for the duration of study therapy.

At 1,800 mg orally every day, 2 of 3 patients experienced grade 3 toxicity deemed probably or definitely related to IPdR. One patient required hospitalization for a community acquired pneumonia (CAP) on Day 7, with associated with cough and grade 3 dehydration. IPdR was possibly a factor in the development of the CAP and IPdR was considered to have contributed to the dehydration. The patient was removed from study (Day 10) with discontinuation of IPdR and interruption of RT (Day 3). The second patient with grade 3 toxicity at 1,800 mg every day presented on Day 22 of IPdR (Day 11 RT) with grade 3 diarrhea in the context of a right perianal abscess requiring incision and drainage. The patient was receiving RT to a left acetabular bone metastasis with a very limited RT field that did not include bowel or sphincter. The abscess was unrelated to study therapy, but it was felt that IPdR probably contributed to the patient's diarrhea. Of note in the context of documented infections, neither patient demonstrated leukopenia or neutropenia at any time. Both patients had resolution of their toxicity without sequelae. The MTD and RP2D based on this study is 1,200 mg orally every day.

Although not the primary objective of the study, all patients had measurable disease at study entry. One patient developed PD of target and nontarget disease on Day 23 of study. Patients that completed study therapy had tumor assessment performed at week 8. Evaluation of target lesions using RECIST criteria revealed 2 complete response (CR), 3 partial response (PR), 9 stable disease (SD), and no PD in these patients (Table 2).

Study patients had metastatic disease and received no systemic therapy for at least 12 weeks (patients were required to be off therapy for at least 4 weeks prior to starting protocol and then received no systemic therapy for the 8 weeks of study therapy). Following assessment at week 8, many patients received additional systemic therapy, and thus, study-related response beyond week 8 could not be determined.

Our preclinical studies showed that the major determinant of IPdR-mediated tumor radiosensitization is the %IUdR-DNA cellular incorporation (40). As such, daily oral IPdR was started 1 week prior to the initiation of RT and continued throughout the RT course to maximize cumulative %IUdR-DNA tumor cell incorporation.

Plasma samples were analyzed for IPdR and its metabolites: IUdR, IUra, and IP. Specific pharmacologic information is not available for IPdR, IP, and IUra, but they were measured because their rates of metabolism influence total exposure to the active species, IUdR (Supplementary Fig. S1; Supplementary Materials and Methods).

Results of the pharmacokinetic analyses are illustrated in Figure 2. Figure 2A presents the mean values in each of 6 groups for plasma concentrations of IUdR (N = 18 patients total). As expected from our previously-published single-dose results (41), concentrations rose relatively quickly during the first hour, and persisted over the 4-hour sampling period. Concentrations increased consistently in each group as doses escalated. Longitudinal comparison across Days 1, 15, and 22 shows excellent consistency, especially for the higher doses that are most relevant to future planning.

Figure 2B presents individual IUdR results for each of the 4 patients who received 1,200 mg, the MTD and RP2D. Interindividual consistency in concentrations was achieved among these patients within each day of sampling, and also across the days of sampling. Patient #14 was lower only on Day 15, and returned to highly consistent levels at Day 22. Patient #18 received 1,800 mg on Day 1, and then 1,200 mg every day Days 2 to 28.

RT is a mainstay of cancer therapy, with close to 60% of patients receiving RT at least once in the course of their treatment (1). RT is used as definitive treatment, in the neo-adjuvant and adjuvant settings to improve and consolidate results of surgical resection, and/or as palliation for cancer-related symptoms. RT can be delivered alone, or more commonly, in combination with chemotherapy and/or surgery. Finally, RT can be administered repeatedly in the course of therapy, either to new areas of disease or, in some cases, re-irradiation to a previously treated site.

The strategy to use a drug to selectively augment the therapeutic efficacy of RT is appealing. However, several issues specific to the mechanism and/or clinical use of radiosensitizers pose formidable challenges with regard to the development process for such agents. In the preclinical setting, the decision to pursue investigation of an oncologic agent is most often based on data from preclinical studies using models that rarely include RT. A selective radiosensitizer with limited or no direct cytotoxic activity fails to elicit responses in such models, and thus is likely to be abandoned. In the clinical arena, current standard-of-care cancer therapy involves multimodality treatment and multiagent chemotherapy that is often administered near the limits of normal tissue/organ tolerance. With potential toxicity and with limited direct antitumor activity, incorporating a radiosensitizer into these regimens is particularly challenging. Finally, pharmaceutical companies fear studying drugs in combination with RT, necessary for the investigation of a potential radiosensitizing agent, as any potential added toxicity could threaten established data regarding tolerability and use of their drug. The strategic guidelines collectively outlined by NCI, RTOG, and leading British investigators to promote the development of radiosensitizers address the distinct issues raised by the study of these agents and highlight the need to make such trials a priority (2, 3).

Several cytotoxic drugs are routinely administered with RT across a range of tumor types because the combined-modality approach has been shown to result in improved locoregional control, progression-free survival, or overall survival compared with single-modality treatment. The therapeutic improvements experienced with the addition of platins (e.g., cisplatin, carboplatin in head and neck, GI, cervix, and lung cancers; refs. 4–6), gemcitabine in pancreatic cancer (12), fluoropyrimidines (e.g., 5-FU, capecitabine in GI cancers; refs. 8–11), and alkylators (e.g., temozolomide in glioma; ref. 13) are welcome options, but they come at a cost of increased acute and late toxicities, thus limiting their therapeutic potential. Agents that target tumor-specific mediators of radiation response/repair would seem to be excellent candidates as radiosensitizing agents. To date, this approach has been largely unsuccessful, for example the inconsistent results and toxicities encountered with the use of EGFR inhibitors to block EGFR-related repair of radiation-induced DNA damage (42–44). How IPdR/IUdR compares in terms of radiosensitization and toxicity to other agents (5-FU and drugs that are not antimetabolites) will ultimately be determined by appropriately-designed clinical trials.

The toxicity profile of IPdR given with concurrent RT was notably favorable in this study. As is typical for phase I trials, the patients in this study were heavily pretreated, and eligibility required that patients have symptomatic metastatic disease necessitating palliative RT. DLT was encountered at 1,800 mg every day, when 2 of 3 patients experienced grade 3 toxicity. The grade 3 diarrhea experienced by 1 patient occurred in the context of an unrelated infection. The other patient developed CAP (with grade 3 dehydration) without evidence of immune compromise (i.e., normal peripheral blood counts). The manageable toxicity encountered in this study is particularly significant in light of the patients' clinical presentations (candidates for palliative RT) and the administration of concomitant, intermediate-dose RT.

The radiosensitization effects of IUdR have been appreciated for decades, but its clinical use is limited by the need for prolonged (24/7 for duration of RT course) infusion and moreover, by systemic toxicities at the doses required for radiosensitization. The extent of radiosensitization caused by IUdR is directly correlated with the amount of IUdR incorporated into tumor cell DNA, increasing with escalating plasma levels of IUdR without a known saturation point (18, 22, 45). In this trial, pharmacokinetic studies indicated that IPdR consistently produces plasma concentrations of the active metabolite, IUdR, within individual patients, and also across the patient population. The exact target range is not defined, but our previous work suggested concentrations of IUdR ≥1 μmol/L for 4 hours would mediate clinically relevant radiosensitization (17–20, 22–25). That goal was clearly met in this study. Results of this trial demonstrate the feasibility and tolerability of PO IPdR administered once daily in combination with an intermediate RT dose, and moreover, tolerable at doses that produce predictable and sustainable plasma IUdR levels that have been associated with clinically relevant radiosensitization.

Investigation of a potential radiosensitizing agent requires not only manipulation of the agent itself (i.e., dose, schedule), but also of the accompanying RT (i.e., dose, schedule, modality, and radiation source). In this study, an intermediate dose (37.5 Gy in 15 2.5 Gy fractions) was delivered by IMRT. Extrapolation of this data to other clinical scenarios, including different patient populations (e.g., tumor type, age, site, and extent of disease, etc.), additional concurrent treatment and alternative RT schemes must be performed with caution. If results of initial phase II studies are encouraging, separate clinical evaluation will be required to assess the toxicity and efficacy of IPdR as a radiosensitizing agent in a range of clinical situations.

The choice to target patients with GI malignancies in this trial was based in part on the expressed interest by the NCI Colorectal Cancer Working Group to investigate the addition of IPdR to standard-of-care therapy for patients with rectal cancer (11). The intermediate RT dose, schedule, and use of IMRT in this study provided results that can be applied to the development of follow-up clinical trials in patients with GI malignancies. Based on the results of this trial, a phase I/II trial of IPdR combining once daily IPdR with standard-of-care RT (45–50.4 Gy over 5–6 weeks) and twice daily capecitabine as neoadjuvant treatment of locally advanced (T₃, T₄, N₁) rectal cancer using pathologic CR rate as the primary endpoint is under consideration.

In this trial, the administration of a single daily oral dose IPdR for 28 consecutive days with concurrent RT was shown to be feasible and tolerable at doses that produce reliable and sustainable plasma IUdR levels ≥1 μmol/L. Systemic toxicities were manageable, especially in light of the heavily pretreated patient population, indicating that the addition of IPdR to standard-of-care, established treatment regimens may be possible. If a phase II trial of IPdR + RT demonstrates a significant improvement in local tumor control with IPdR + RT, the impact could decrease rates of both local and distant failure, and perhaps permit lower RT dose for pediatric patients, elderly patients, and other patients for whom the toxicity of RT poses particular risk. Furthermore, the mechanism of IPdR/IUdR radiosensitization should theoretically extend to other forms of therapeutic RT including particle therapy and hypo- and hyperfractionation schedules. Thus, the results of this trial strongly support the investigation of the efficacy of IPdR + RT in phase II studies.

T. Kinsella and S. Wiersma are employees of and hold ownership interest (including patents) in EMEK, Inc. No potential conflicts of interest were disclosed by the other authors.

Conception and design: T. Kinsella, H. Safran, C. Kunos, J.M. Collins

Development of methodology: T. Kinsella, H. Safran, L.W. Anderson, K.D. Hill, C. Kunos, J.M. Collins

Acquisition of data (provided animals, acquired and managed patients, provided facilities, etc.): T. Kinsella, T. DiPetrillo, A. Schumacher, J. Vatkevich, L.W. Anderson, K.D. Hill, C.A. Kunos

Analysis and interpretation of data (e.g., statistical analysis, biostatistics, computational analysis): T. Kinsella, H. Safran, T. DiPetrillo, L.W. Anderson, C. Kunos, J.M. Collins

Writing, review, and/or revision of the manuscript: T. Kinsella, H. Safran, S. Wiersma, T. DiPetrillo, C. Kunos, J.M. Collins

Administrative, technical, or material support (i.e., reporting or organizing data, constructing databases): S. Wiersma, A. Schumacher, J. Vatkevich, C. Kunos, J.M. Collins

Study supervision: T. Kinsella, H. Safran, A. Schumacher, J. Vatkevich, C. Kunos, J.M. Collins

Other (assisted the principal investigator with administrative support to document protocol amendments and memos): K. Rosati

This project has been funded in whole or in part with federal funds from the NCI, NIH, under contracts nos. HHSN261200800001E and HHSN261201400013C (Small Business Innovation Research contract through Shuttle Pharmaceuticals, Inc.). The content of this publication does not necessarily reflect the views or policies of the Department of Health and Human Services, nor does mention of trade names, commercial products, or organizations imply endorsement by the U.S. Government. Support for this study was also provided by the DBJ Foundation and the University Radiation Medicine Foundation (to T. Kinsella).

BrUOG Co-investigators: R. Breakstone, K. Leonard, K. Mantripragada, M. LeGolvan

The patients who have participated in the clinical trials of IPdR.

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.