Noninvasive Molecular Imaging Sheds Light on the Synergy between 5-Fluorouracil and TRAIL/Apo2L for Cancer Therapy

Apoptosis (programmed cell death) is a highly regulated cellular process that serves an essential physiologic role in the removal of unwanted cells and is essential for proper development of living organisms (1–3). Activation of the programmed cell death pathway can also serve a protective role by eliminating damaged cells, which have lost proper function, to combat and/or prevent disease. As such, aberrations in apoptotic signaling have been implicated in numerous diseases, most notably cancer, as the evasion of apoptosis is considered to be one of several defining hallmarks of cancer pathogenesis (4).

Within living cells, the regulation of apoptosis involves a complex balance of signaling pathways that ultimately interact with the apoptotic pathway by instigating either proapoptotic or antiapoptotic signals leading to activation or suppression of cell death. As such, defects in any of these signaling pathways can have deleterious results leading to resistance to cell death and eventual expansion of a population of neoplastic cells. In response, novel anticancer therapies are in active development to overcome these defects and elicit cellular death (5–8). An impeding hurdle in drug development is the difficulty in obtaining reliable and quantitative preclinical assessment of overall therapeutic value especially in the case of targeted therapies such as proapoptotic agents. Currently, standard procedures for in vivo drug validation rely on cumbersome volumetric measurements with changes in tumor growth kinetics serving as the determinant of therapeutic benefit. However, volumetric assessments can be highly subjective as well as misleading, especially in targeted therapeutics that may not cause overt tumor regression. Attempts to understand the changes in molecular mechanisms elicited by these therapeutic agents typically require invasive techniques using ex vivo biochemical assays. Although histologic staining of tissue sections can provide valuable information, these methods can be time-consuming, which require subjective observers as well as the use of a large animal numbers serially sacrificed at each time point of interest. Therefore, the development of a quantitative technique to detect apoptosis noninvasively could provide immense advantages for evaluating therapeutic strategies in vivo.

In a previous report, we described an imaging platform capable of noninvasive and quantitative assessment of apoptotic activity in vivo through the use of bioluminescence imaging (9). To achieve this, a recombinant reporter molecule (ERLucER) was generated by fusing the estrogen receptor regulatory domains at the NH₂ and COOH terminus of firefly luciferase with caspase-3 cleavage sites (DEVD) inserted between the two domains. When tumors expressing the hybrid molecule were generated, minimal reporter activity was detectable under normal conditions; however, when treated with the experimental death-inducing peptide tumor necrosis factor α–related apoptosis–inducing ligand (TRAIL), a significant increase in luciferase activity was observed and later validated to be indicative of apoptosis in vivo. In the present study, we aimed to further show the sensitivity of this reporter molecule by evaluating the antitumoral effects of a combinatorial therapy involving TRAIL and 5-fluorouracil (5-FU). Numerous studies have previously shown that TRAIL sensitivity in cell lines can be modulated through concomitant treatment with an array of genotoxic agents (10, 11), and that this phenomenon occurs in a p53-dependent manner (12, 13). However, conventional means to evaluate the synergistic nature of conventional DNA-damaging agents and TRAIL are limited by the need to perform serial biochemical analyses on tumor samples. As such, the ability to both determine the appropriate timing and scheduling of treatments and to evaluate the kinetics of tumor response to multiple agents becomes prohibitively difficult. Therefore, we used our hybrid reporter molecule to dynamically observe apoptotic activity in vivo as a means to overcome these hurdles by providing real-time readout of this biological process.

Although bioluminescence imaging is an evolving technique for preclinical evaluation of biological processes, it is not readily transferable to clinical practice. In contrast, diffusion magnetic resonance imaging (MRI) has been consistently shown to provide an early and quantitative assessment of tumor response that is predictive of overall therapeutic outcome (12, 14–19). By tracking changes in the molecular motion of water [designated as the apparent diffusion coefficient (ADC)], diffusion MRI gives insights into changes in tumor cellularity and membrane permeability that can occur well before changes in tumor volume or other MRI variables (T1, T2, fast fluid–attenuated inversion recovery, or contrast enhancement; refs. 18, 20). An additional advantage of diffusion MRI is the direct clinical translatability of the technology, which could be used to monitor response to therapies in clinical trials (21, 22). In this study, we explored the use of diffusion MRI as a surrogate marker for tumor response resulting from a predominantly proapoptotic therapy as evidenced by our bioluminescence reporter system.

Stable cell lines. Stable cell lines expressing the ER-Luc-ER construct were generated as previously described (9). Briefly, the expression plasmid was stably transfected into D54 human glioma cells using Fugene (Roche Diagnostics, Basel, Switzerland), and the resulting stable clones were selected using selection medium containing 200 μg/mL G418 (Invitrogen, Carlsbad, CA) and maintained throughout isolation and characterization. Specific clones were identified and selected for further study based on expression levels and activity of the recombinant protein.

In vivo studies. D54 human glioma cells constitutively expressing the ERLucER reporter molecule were grown as monolayers in RPMI supplemented with 10% FCS and 200 μg/mL G418 in a 95%/5% air/CO₂ atmosphere. S.c. D54 tumors were induced in CD-1 nu/nu athymic mice (Charles River Laboratories, Boston, MA) by implantation of 10⁷ cells suspended in 0.2 mL of 1:1 serum-free RPMI/Matrigel (BD Biosciences, San Jose, CA). When tumors reached ∼40 to 80 mm³, treatment protocols were initiated with mice subdivided into four groups with one serving as the saline-treated control group, whereas the other three groups were treated with 5-FU, TRAIL, or a combination of 5-FU and TRAIL (5-FU + TRAIL). The 5-FU group received two i.p. injections of 60 mg/kg 5-FU (Sigma, St. Louis, MO) at days 0 and 5. The TRAIL-treated group received daily tail vein i.v. injections of 4 mg/kg TRAIL for 10 days beginning day 1 through 10. The TRAIL used is a truncated His-tagged version and was generated in Escherichia coli and purified as previously described (12). Last, the combined treatment group received both 5-FU and TRAIL at the aforementioned dosing schedules and was initiated with 5-FU (day 0) being administered 1 day before TRAIL (day 1) treatment. Tumor volumes were obtained using MRI and were measured every 2 to 3 days until cessation of the experiment. Specific tumor growth delay was calculated as the difference in days for treated tumors to reach four times the initial volume compared with control animals. Differences in treatment groups were determined using Student's t test (one-tailed equal variance).

Bioluminescence imaging. Bioluminescence imaging was initiated before drug administration for baseline values and done throughout the course of the experiment. For bioluminescence imaging, mice were anesthetized with a 2% isofluorane/air mixture and given a single i.p. dose of 150 mg/kg d-luciferin (Promega, Madison, WI) in normal saline. Subsequently, animals were removed from anesthesia and allowed to awaken. Animals were then reanesthetized ∼8 min after administration of d-luciferin, and images were acquired ∼10 to 12 min after administration of d-luciferin. For image acquisition, a cryogenically cooled charge coupled device camera system (Xenogen, Alameda, CA), with a nose cone isofluorane delivery system and heated stage for maintaining animal body temperature, was used. The variables used for luminescence imaging were f/stop 1, bin 2, field of view 10 cm, and time 30 s. Results were analyzed using Living Image software provided with the Xenogen imaging system. A gray-scale body image was collected followed by luminescence acquisition, which was overlaid as a pseudocolor image representing the spatial distribution of the detected photons emitted from the tumor. Signal intensity was quantified as the sum of all detected photon counts within a uniform region of interest manually placed over the tumor site during data postprocessing.

Diffusion MRI. For MRI examination, mice were anesthetized with a 2% isofluorane/air mixture and maintained at 37°C inside a 7-T Varian Unity Inova imaging system as previously described (17). A single-slice, gradient-echo sequence was used to confirm proper animal positioning and to prescribe subsequent imaging. An isotropic, diffusion-weighted sequence was used with two interleaved b-factors (Δb = 1,148 s/mm²) and the following acquisition variables: TR/TE 3,500/60 ms, 128 × 128 matrix, and a 3-cm field of view. Thirteen 1-mm-thick slices separated by a 0.25-mm gap were used to cover the whole tumor. The z gradient first moment was zeroed to reduce the dominant source of motion artifact. To further reduce motion artifact, a 32-point navigator echo was prepended to each phase-encode echo. The phase deviation of each navigator echo relative to their mean was subtracted from the respective image echoes before the phase-encode Fourier transform. The low b-factor images were essentially T2 weighted to allow tumor volume measurements. Images were acquired before treatment and at 2- to 3-day intervals thereafter. Isotropic ADC maps were calculated for each image set and ADC pixel value histograms were generated from tumor regions of interest combined across slices. Tumor volume was assessed on late-echo images in which the tumor seems hyperintense. These images were acquired as part of the diffusion-weighted scan (low b-factor images, TR/TE 3,500/60 ms) or in a separate standard T2-weighted series (TR/TE 3,000/80 ms). The tumor boundary was manually defined on each slice using a region-of-interest tool, then integrated across slices to provide a volume estimate.

Immunostaining. Paraffin sections were cut on a microtome and heated for 20 min at 65°C. Slides were deparaffinized in xylene with three changes of 2 min each, then rehydrated through an alcohol gradient of 2 min each (100% alcohol, 95% alcohol, 70% alcohol). Antigen retrieval was done by heating the slides in citrate buffer (pH 6.0) for 10 min in a microwave oven. Tissues were then blocked in PBS-T supplemented with 5% normal donkey serum for 1 h. For p53 immunohistochemistry, a p53 antibody (DAKO Corp., Carpinteria, CA) was used at a dilution of 1:50 and the secondary label was achieved using the EnVision+ labeled Polymer (DAKO). After buffer rinse, the sections were incubated with 3,3′-diaminobenzidine chromagen (DAKO) for 5 min, quickly rinsed in water, and subsequently hematoxylin counterstained. Slides were then redehydrated using an increasing gradient of alcohol, rinsed in xylene, and subsequently overlaid with a coverslip. Activation of caspase-3 was monitored using an antibody specific for the cleaved form of caspase-3 (Cell Signaling Technology, Beverly, MA) at 1:100 dilution. A cy3-coupled anti-rabbit secondary antibody was used at a dilution of 1:400 (Jackson Immunoresearch, West Grove, PA). The slides were then costained with 1 μg/mL 4′,6-diamidino-2-phenylindole, mounted, and visualized under a fluorescence confocal microscope Nikon Eclipse TE2000-U (Nikon, Melville, NY). All fluorescence images were acquired using Metamorph software (Molecular Devices Corporation, Sunnyvale, CA) under the same exposure times and images were exported as TIFF files. For quantitation, the average pixel intensity of the active caspase-3 signal in nine randomly selected fields was measured using Metamorph software and is expressed as arbitrary fluorescence units (AFU).

5-FU leads to synergistic enhancement of TRAIL-induced apoptosis as detected by bioluminescence imaging. In this study, we chose to use a noninvasive reporter for caspase-3 activity as a surrogate for apoptosis (9) to dynamically monitor the apoptotic response of D54/ERLucER xenografts to various therapeutic regimens of 5-FU, TRAIL, or 5-FU + TRAIL. Bioluminescence imaging was used to monitor reporter activity throughout the course of therapy to assess temporal changes in caspase-3 activity. In Fig. 1, bioluminescence images of representative mice from each group for day 0, 1, and 6 of the treatment course are shown as a qualitative demonstration of the different patterns of signal induction in response to therapy. The representative animals from control (n = 4) and 5-FU–treated (n = 4) groups exhibit only slight deviations in bioluminescence, which is consistent with a lack of caspase-3 activation. In contrast, a marked increase in bioluminescence was observed on day 1 after initiation of therapy in animals from the TRAIL (n = 5) and 5-FU + TRAIL (n = 5) groups. However, at day 6, a decrease in bioluminescence was detected in the TRAIL alone–treated tumor although daily treatments of TRAIL were still being administered. In contrast, the 5-FU + TRAIL–treated tumor had a persistent temporal activation of apoptosis in which a second treatment with 5-FU, on day 5, actually led to an observable increase in bioluminescence signal after TRAIL treatment on day 6 compared with day 1.

To assess differences in apoptosis between each group, the mean percentage change of the maximal luminescence signal from each day was calculated and plotted (Fig. 2). As expected, control tumors had little change in bioluminescence throughout the course of the experiment with a slight increase in bioluminescence near the end. Similarly, the 5-FU group displayed minimal deviations in caspase-3–dependent bioluminescence throughout the course of therapy, showing that 5-FU had minimal proapoptotic effects. However, TRAIL treatment showed an increase of almost 400% change within the 1st day of injection. Interestingly, although daily injections continued for 10 days, a gradual drop in signal was observed, indicating a decrease in caspase-3 activity, which from day 5 onward was not statistically different from the control group (P > 0.1 on day 5) although the tumors continued to receive daily TRAIL injections of days 6 to 10. In contrast, when tumors were treated with a combination treatment of 5-FU and TRAIL, a sustained increase in bioluminescence was observed throughout the first 5 days. Bioluminescence further increased when the second 5-FU dose was administered on day 5, leading to an ∼500% higher apoptotic activity in 5-FU + TRAIL–treated animals compared with TRAIL alone for the remainder of therapy. Thus, the combination of 5-FU and TRAIL not only produced an enhanced caspase-3–dependent bioluminescence response but also was sustained throughout the course of the experiment, which was in direct contrast to TRAIL alone therapy in which apoptotic activity progressively diminished. Importantly, TRAIL alone only generated a transient induction of caspase-3–dependent bioluminescence that diminished with each successive TRAIL treatment, suggesting that tumors treated with single-agent TRAIL rapidly lost responsiveness to therapy.

Immunostaining of p53 and active caspase-3 correlate with bioluminescence imaging of apoptosis. To validate that the increased caspase-3–dependent bioluminescence observed in the 5-FU + TRAIL–treated tumors corresponded with a concurrent increase in apoptosis, an equivalent set of tumors was treated using the same experimental protocol for histologic analysis. With numerous studies suggesting that sensitization of TRAIL-induced apoptosis occurs through a p53-dependent pathway (10, 12, 13, 23), immunohistochemical analyses (Fig. 3) were done on tumor sections to ascertain therapy-induced changes in p53 activity. As expected, control tumors exhibited basal levels of p53 (Fig. 3) throughout the course of the experiment. Treatment with TRAIL alone did not elicit detectable changes in p53 levels at day 1 or 6. However, when 5-FU was administered in the 5-FU and 5-FU + TRAIL groups, a dramatic increase in p53 levels was revealed at days 1 and 6 corresponding to the dosing schedule of two bolus injections of 5-FU at days 0 and 5.

Because caspase-3 activation is a central component of the cell death cascade and the activator of the hybrid bioluminescence molecule used in these studies, immunofluorescence was done to ascertain the status of caspase-3 activity within tumor sections (Fig. 4A). The results for each group were quantified and expressed as AFU in Fig. 4B. As expected from our bioluminescence imaging experiments, 5-FU sections exhibited minimal staining for the active form of caspase-3 throughout the course of the experiment in which day 1 (301 ± 84 AFU; P < 0.0001) and day 6 (320 ± 64 AFU; P < 0.001 versus control) sections exhibited minimal changes in caspase-3 activity. However, sections from TRAIL alone–treated (459 ± 141 AFU) and 5-FU + TRAIL–treated (623 ± 164 AFU) tumors revealed a dramatic increase in caspase-3 activity on day 1 compared with pretreatment tumor samples (TRAIL, P < 0.00008; 5-FU + TRAIL, P < 0.0000025) as well as the control group (TRAIL, P < 0.00001; 5-FU + TRAIL, P < 0.000002). However, caspase-3 activity in the TRAIL tumor was greatly diminished at day 6 (272 ± 69 AFU; P < 0.0006 versus day 1), which confirmed earlier revelations that the observed initial induction of apoptosis by TRAIL was followed by diminishing bioluminescence throughout the remainder of the therapeutic protocol. In contrast, 5-FU + TRAIL samples showed increased levels of active caspase-3 even at day 6 (928 ± 65 AFU; P < 0.000001 versus day 1), which is consistent with the bioluminescence imaging findings showing that concomitant 5-FU and TRAIL resulted in sustained elevated levels of caspase-3 activity throughout the course of therapy.

5-FU and TRAIL combination therapy provides enhanced tumor growth delay. Having established the enhanced induction of apoptosis with concomitant 5-FU and TRAIL therapy in vivo using both bioluminescence imaging and histology, we next investigated the antitumoral effects in vivo using standard volumetric measurements. For each group, the average percentage change in tumor volume was determined and plotted in Fig. 5A. In addition, specific growth delays (determined as the difference between the treated groups and control in the number of days required to reach 200% original volume) were also calculated and summarized in Fig. 5B. Saline-treated control mice (n = 4) exhibited a tumor doubling time of ∼9 days. In 5-FU–treated mice (n = 4), tumor growth curves revealed a slight decrease in tumor growth; however, the specific growth delay (1.7 days) did not meet statistical significance when compared with controls (P > 0.2). When mice were treated with TRAIL (n = 5), a more apparent decrease in tumor growth rate was detected with an estimated specific growth delay of 4 days (P < 0.005 versus control). However, tumor growth inhibition from TRAIL treatment was only observed in the first 2 days (Fig. 4A), as the slope of the growth curve from days 4 to 14 seemed to be similar to the growth curve from the control group. This is consistent with bioluminescence imaging findings in which a transient spike in caspase-3–dependent bioluminescence observed in the first several days of TRAIL treatment signaled initial efficacy of therapy, which subsequently dissipated despite continuing daily treatment with TRAIL.

Finally, in 5-FU + TRAIL–treated mice (n = 5), an appreciable growth inhibition was revealed as evidenced by tumor volumes remaining constant throughout the entire 10 days of therapy. However, shortly after the completion of therapy, the tumors resumed a growth rate with slope similar to that initially observed in untreated animals. This eventually led to a specific growth delay of ∼9 days, which was statistically significant compared with each agent alone (5-FU, P < 0.05; TRAIL, P < 0.05 and control P < 0.005).

Concomitant 5-FU and TRAIL therapy led to enhanced therapeutic effect as detected by diffusion MRI. Because bioluminescence, histology, and tumor volumes showed enhanced treatment efficacy in the 5-FU + TRAIL group, we further investigated the ability of diffusion MRI to serve as an early biomarker for this response. The mean maximal change in diffusion due to each therapeutic regimen was calculated and summarized in Fig. 6. Control (n = 4) animals exhibited only a 12% increase in ADC throughout the experiment, which likely reflected areas of central necrosis (18). 5-FU (n = 4) also failed to produce a dramatic change in ADC with an increase of 12%, which was not different when compared with control animals (P > 0.5 versus controls). The TRAIL-treated group (n = 5) exhibited a modest increase in ADC of ∼20%, although this did not achieve statistical difference compared with control (P > 0.1) or 5-FU (P > 0.1). This is consistent with the observed initial increase in caspase-3–dependent bioluminescence and the early slight change in the level of active caspase-3, which both dissipated over the course of treatment. However, when mice were treated with both 5-FU and TRAIL (n = 5), a dramatic 32% increase in diffusion was revealed, which was statistically significant versus TRAIL (P < 0.05), 5-FU (P < 0.02), and control (P < 0.002) groups. This finding correlated with the apoptosis-imaging data signifying that concurrent therapy of 5-FU and TRAIL indeed induced greater tumor response as shown by diffusion MRI.

The role of apoptosis in oncology is clear (1, 4) in which the decoupling of cellular proliferation pathways and death pathways may be a major determinant of both pathogenesis and response to treatment. As such, strategies to reconnect these intricate networks have become of intense interest, driving the development of novel therapies aimed at restoring appropriate death signaling. For example, TRAIL, an apoptosis-inducing member of the tumor necrosis factor family of genes, has garnered a great deal of attention. Through mechanisms that are not entirely clear at this time, TRAIL seems to have a highly selective ability to kill many cancer cells with the added feature of being nontoxic to normal differentiated tissues (24). With this influx of apoptosis-modulating strategies in preclinical and clinical development, there is a growing need for the efficient evaluation of these experimental therapies.

Tumor growth measurements remain the gold standard in assessing therapeutic efficacy; however, this technique is both time-consuming and easily susceptible to human error. In a previous report, we described an imaging strategy that provided a sensitive and quantitative approach for evaluating caspase-3–dependent apoptotic activity noninvasively (9). In the present study, we further show the sensitivity of this reporter system and its ability to provide real-time monitoring of therapeutic effect for evaluation of different treatment regimens. More importantly, this bioluminescence reporter provides a unique ability to follow the temporal pattern of caspase-3 activity in response to treatment, which provided insight into the synergistic interaction between 5-FU and TRAIL. Traditional volumetric measurements for assessing tumor response would have indicated that TRAIL was only modestly effective in this treatment regimen, despite our observations of significant activation of caspase-3 observed at day 1 after treatment. Further, conventional assessment of response would likely not have revealed the loss of tumor response to TRAIL-induced apoptosis after subsequent treatments, which could provide an explanation for the minimal overall benefit from TRAIL treatment.

Using the cytotoxic peptide TRAIL in conjunction with 5-FU, a rapid in vivo assessment of treatment synergy was achieved using the apoptosis-imaging platform. Although 5-FU was ineffective in inducing apoptosis, the combination of 5-FU and TRAIL led to prolonged elevated apoptotic activity that was sustained throughout the 10 days of therapy (Fig. 2). This is consistent with other reports demonstrating that genotoxic agents can sensitize tumor cells to TRAIL (10, 23, 25, 26) and that this mechanism is mediated through p53 (12, 13). Interestingly, in the 5-FU + TRAIL group, a second injection of 5-FU on day 5 led to a further increase in apoptotic response, detected by bioluminescence imaging, as TRAIL therapy continued from day 6 to 10. Presumably, an increase in apoptotic response should result in greater growth inhibition, although a difference in growth velocity was not observed when comparing tumor volumes from day 1 to 5 and day 6 to 10 in the 5-FU + TRAIL group. There are numerous plausible explanations for this phenomenon. Possibly, given that a high level of apoptotic activity was already achieved using 5-FU + TRAIL, perhaps an even greater threshold of apoptotic activation needs to be reached to translate into further growth inhibition and tumor regression. However, this is purely speculative and further studies are required to validate this notion.

When TRAIL was given alone, only a transient induction of apoptosis was produced, which completely receded by day 4 and remained diminished throughout the experiment. Presumably, this perceived loss of apoptosis activity suggested that the tumors had reduced sensitivity to TRAIL. A plausible explanation for this observed phenomenon is that TRAIL therapy eradicated a responsive subpopulation of cells, resulting in a tumor repopulated by a remainder of cells insensitive to TRAIL. Notably, because TRAIL-treated tumors exhibited continued growth throughout treatment (Fig. 5A), the loss in bioluminescence as therapy progressed was most likely not an artifact, in which a lack of caspase-3 activity was due to the tumor population consisting of dead cells, but rather a loss in TRAIL sensitivity. Conversely, 5-FU + TRAIL treatment showed a synergistic enhancement in apoptosis activation as evidenced by sustained elevated apoptotic activity throughout the entire course of treatment as seen by bioluminescence. These results taken together validated the advantages of this imaging platform in which we were able to detect in real-time the emergence of TRAIL resistance and, more importantly, the synergistic interaction of TRAIL and 5-FU achieved through quantitative analysis of the apoptotic response in vivo.

With our apoptosis-imaging and histology data indicating enhanced induction of apoptosis through concomitant 5-FU and TRAIL therapy, we used diffusion MRI as a means to detect and quantify differences in cellular response to treatment for each therapeutic regimen. Diffusion MRI exploits the molecular motion of water as a biomarker for cellularity and has been shown to be a reliable indicator of tumor response to therapy both in preclinical and clinical studies (12, 14, 21, 22). Moreover, a recent study highlighted the sensitivity of this approach in which diffusion MRI was able to detect in real-time the emergence of drug resistance during therapy (15). As shown in Fig. 6, treatment with either 5-FU or TRAIL alone failed to produce a significant increase in ADC compared with controls. Presumably, tumor response to TRAIL therapy was below the threshold of detection by diffusion MRI, correlating with a previous study that showed a similar finding (12). This was not entirely surprising given that our apoptosis-imaging data (Fig. 2) showed that TRAIL treatment produced a rapid, yet transient response that quickly degraded by day 4, which likely explains the lack of an overall therapeutic effect. However, concomitant treatment with 5-FU and TRAIL indeed produced a significant increase in ADC, indicating that the combinatorial approach was much more efficacious compared with either agent alone. Again, these findings corroborated the bioluminescence imaging and histologic findings confirming that combination 5-FU and TRAIL was indeed more efficacious. Taken together, these findings further show that diffusion MRI can serve as a surrogate marker for noninvasive and rapid assessment of tumor response to a predominantly proapoptotic therapy. This is of particular importance because TRAIL-based therapies are progressing into phase I/II clinical trials, and the findings in this study indicate that diffusion MRI may be a useful tool to better understand the value of these or other experimental therapies in clinical trials.

Finally, this study further provides the proof of principle that the apoptosis bioluminescence reporter molecule serves as a sensitive readout that provides a unique advantage for real-time assessment of a biological process in vivo. However, this bioluminescence approach is not limited to imaging apoptosis, in which exchanging the caspase-3 cleavage site with another specific cleavage sequence could extend this imaging strategy for monitoring other proteases relevant to disease. Another potential consideration is adapting this reporter silencing/reactivation approach for other imaging modalities by substituting the enzyme with other known reporter molecules.