Homozygous deletion of the methylthioadenosine phosphorylase (MTAP) gene is a frequent event in a wide variety of human cancers and is a possible molecular target for therapy. One potential therapeutic strategy to target MTAP-deleted tumors involves combining toxic purine analogues such as 6′-thioguanine (6TG) or 2′-fluoroadenine (2FA) with the MTAP substrate 5′-deoxy-5′-methylthioadenosine (MTA). The rationale is that excess MTA will protect normal MTAP+ cells from purine analogue toxicity because MTAP catalyzes the conversion of MTA to adenine, which then inhibits the conversion of purine base analogues into nucleotides. However, in MTAP tumor cells, no protection takes place because adenine is not formed. Here, we examine the effects of 6TG and 2FA in combination with MTA in vitro and in vivo. In vitro, MTA protected against both 6TG and 2FA toxicity in an MTAP-dependent manner, shifting the IC50 concentration by one to three orders of magnitude. However, in mice, MTA protected against toxicity from 2FA but failed to protect against 6TG. Addition of 100 mg/kg MTA to 20 mg/kg 2FA entirely reversed the toxicity of 2FA in a variety of tissues and the treatment was well tolerated by mice. The 2FA+MTA combination inhibited tumor growth of four different MTAP human tumor cell lines in mouse xenograft models. Our results suggest that 2FA+MTA may be a promising combination for treating MTAP-deleted tumors.

Significance: Loss of MTAP occurs in about 15% of all human cancers; the MTAP protection strategy presented in this study could be very effective in treating these cancers. Cancer Res; 78(15); 4386–95. ©2018 AACR.

A quarter century ago, Toohey first recognized that certain murine malignant hematopoietic cell lines lacked methylthioadenosine phosphorylase (MTAP) activity (1). MTAP is a key enzyme in the purine and methionine salvage pathways that catalyzes the conversion of the polyamine byproduct 5′-deoxy-5′-methylthioadenosine (MTA) into the purine base adenine and the sugar methylthioribose-1-phosphate. MTAP is ubiquitously expressed at high levels in all tissues in the human body. The gene encoding human MTAP is located on chromosome 9p21, about 80 kb toward the telomere from the CDKN2A/ARF region (2, 3). Homozygous deletion of MTAP is frequent in a large number of different human cancers including leukemia, lymphoma, lung, pancreas, squamous cell, billiary tract, brain, bone, breast, prostate, bladder, mesothelioma, endometrial, melanoma, gastrointestinal, and neuroendocrine (4–28). It is also important to note that deep sequencing suggests that homozygous deletion of the 9p21 region appears to be an early event in tumor evolution (29), and IHC staining for MTAP also suggests a distinct lack of heterogeneity (10, 30, 31). Taken together, these findings suggest that a therapy that specifically targets MTAP cells could be useful in treating a wide variety of different types of cancer. Recently, three different groups have performed synthetic lethal screens and discovered that MTAP cells are more sensitive to downregulation of protein arginine N-methyltransferase 5 (PRMT5; refs. 32–34). However, when pharmacologic inhibitors of PRMT5 were used, they seemed to inhibit cell growth with either no (34) or little relationship to MTAP status (32). In addition, because PRMT5 is essential for cell viability, it is unclear whether pharmacologic inhibition of PRMT5 can be used without significant side effects.

Here, we explore a different strategy to target MTAP cells involving a combination of purine analogues and MTA. Purine analogues (PA), such as mercaptopurine or 6-thioguanine, are among the oldest chemotherapy agents (35). These analogues are actually prodrugs, which must be converted to nucleotides to be active. This process occurs through the action of specific phosphoribosyl transferase enzymes that transfer a ribose and phosphate from phosphoribosylpyrophosphate (PRPP) to the base. Once activated, these drugs interfere with DNA replication and cellular metabolism by inhibiting enzymes that normally bind nucleotides. The major limitation with this class of drugs is that they target all dividing cells, not just cancer cells. At a clinical level, this means the physician has a limited “therapeutic window” in which to use these drugs, balancing the killing of tumor cells with the toxicity to the patient. In theory, if the effect of these agents on normal tissue could be ameliorated, much higher doses could be used and this would likely increase their clinical effectiveness.

One way to protect cells from the toxicity of purine analogues is to prevent their conversion to nucleotides by competition for PRPP (Fig. 1). When MTA is combined with purine analogues in the presence of MTAP, excess MTA is converted to adenine, which is then reacted with PRPP to make adenosine 5′-monophosphate (AMP). This reaction is catalyzed by adenine phosphoribosyltransferase (APRT). Thus, in the presence of high concentrations of MTA, PRPP levels may become depleted and PAs cannot not be converted to 5′-monophosphate form of the purine analogue (PAMP). However, in tumor cells lacking MTAP, no excess adenine is produced, resulting in increased production of PAMP and increased cell death.

Figure 1.

MTAP protection strategy. In MTAP+ cells, MTA is converted to adenine, which is then used to synthesize AMP by the action of APRT, which transfers the ribose and phosphate from PRPP. PAs must also be converted to nucleosides (PAMP) by the transfer ribose and phosphate from PRPP (although the exact phosphoribosyltransferase used can vary depending on the analogue). High concentrations of MTA inhibit the conversion of PA to PAMP by either competition for PRPP, or competition for APRT. In MTAP cells (bottom), MTA is not converted to adenine, so no competition takes place, resulting in increased production of toxic PAMP.

Figure 1.

MTAP protection strategy. In MTAP+ cells, MTA is converted to adenine, which is then used to synthesize AMP by the action of APRT, which transfers the ribose and phosphate from PRPP. PAs must also be converted to nucleosides (PAMP) by the transfer ribose and phosphate from PRPP (although the exact phosphoribosyltransferase used can vary depending on the analogue). High concentrations of MTA inhibit the conversion of PA to PAMP by either competition for PRPP, or competition for APRT. In MTAP cells (bottom), MTA is not converted to adenine, so no competition takes place, resulting in increased production of toxic PAMP.

Close modal

Previously published work by two different groups shows that this general strategy works well in cell culture (36, 37). However, the key question of whether MTA can protect against PA toxicity in vivo has not been definitively explored. There are two reports in the literature suggesting that MTA can protect against 6TG toxicity in mice, but these studies used very few animals and are not well described (20, 38). Here, we have performed a series of experiments examining the ability of MTA to protect against toxicity to two different purine analogues, 6TG and 2′-fluoroadenine (2FA). Surprisingly, while we see excellent MTAP/MTA-mediated protection of both compounds in cell culture, we only see MTAP/MTA-mediated protection for 2FA in mice. Furthermore, the combination of 2FA and MTA inhibits tumor growth in several different MTAP xenograft models. Our findings suggest that 2FA+MTA may be useful in the treatment of MTAP cancers.

Chemicals and drug preparation

For cell culture studies, 6TG (Sigma), 2FA (Oakwood Chemical), and MTA (Sigma) were dissolved directly in cell culture medium. For mouse studies, 6TG was dissolved at a concentration of 2.7 mg/mL in 1% carboxymethylcellulose (Sigma) in filter-sterilized water. For 6TG/MTA, MTA was added to the 6TG carboxymethylcellulose solution at a concentration of 3.6 mg/mL. 2FA was dissolved at a concentration of 0.36 mg/mL in the presence of equal normality sulfuric acid in a 1% carboxymethylcellulose solution. MTA was added to the 2FA solution at a concentration of 1.8 mg/mL.

Cell lines

Cell lines, along with MTAP status, used for this study are listed in Supplementary Tables S1 and S2. All cell lines were originally obtained either from the ATCC or were created in the Fox Chase Cancer Center Cell Culture Facility and have been described previously (37), with the exception of the NIH3T3 cells that have had MTAP inactivated by CRISPR (see below). Cells were grown in 2 mmol/L glutamine, 100 μg/mL penicillin, 100 μg/mL streptomycin, 10% FBS, and 250 μg/mL G418 for no more than 5 passages after thaw. Cells were tested for Mycoplasma at the time of thaw. No cell line authentication was performed by our laboratory.

We generated Mtap deletion in NIH3T3 cells by CRISPR/Cas9 using a method in Bauer and colleagues (2015). Two sets of single guide RNAs (sgRNA) were designed to delete Mtap exons 5 and 6 (sgRNA-A: CACCGGGCAAAACGGTTCAGCCAT, sgRNA-A-rc: AAACATGGCTGAACCGTTTTGCCC, sgRNA-B: CACCGTTTCCTTTGCATTTGTCTCG, sgRNA-B-rc: AAACCGAGACAAATGCAAAGGAAAC). For CRISPR cloning, phosphorylated oligos were ligated into Bbs 1 site of eSpCas9 (1.1)-1 vector and transformed into DH5α cells. Colonies were verified by sequence using U6 promoter primer (CGTAACTTGAAAGTATTTCGATTTCTTGGC). Each CRISPR pair was cotransfected with GFP expression plasmid (pAcGFP1-N1) into NIH3T3 cells by electroporation using Amax Cell Line Nucleofactor Kit R (Lonza). Top, approximately 3%, of GFP-positive cells were sorted by FACS and plated individually to grow clone. Genomic DNAs from parental and sorted cells were isolated and validated Mtap deletion by PCR (Qiagen HotStar Taq Master Kit) using deletion screening primers (MTAP del-F: GTCTTTGGAGGGACAGCAAG and MTAP del-R: ACCCATGTGGCGTAAGAGTC). Selected clones were further evaluated for Mtap deletion by sequencing and MTAP activity.

IC50 studies

For IC50 studies, cells were plated out at a density of 3,000 cells per well in a 96-well cell culture plate. Wells were exposed to either nothing or drug at 2-fold increasing concentrations over a 128-fold range. Each dose was tested in triplicate. Plates were incubated for 48 hours. Growth was measured by MTT assay (Cell Titer 96, Promega). IC50 values were then calculated using GraphPad Prism 5.0 software.

Mixing experiment using CRISPR MTAPΔ cells

Parental NIH3T3 cells and CRISPR MTAPΔ cells were mixed either 1:1 or 4:1 ratio and then treated with either none or 2FA (1.25 μmol/L)+ MTA (10 μmol/L). After 70 hours of incubation, cells were harvested and extracted genomic DNAs were performed allele-specific PCR using specific primers either for wild-type MTAP (GTCTTTGGAGGGACAGCAAG and CACACATTCAGGTCCACTGC, 300 bp PCR product) or for MTAPΔ allele detection (TGCCAGAGGATCGGGGCCTTG and ACCCATGTGGCGTAAGAGTC, 706 bp PCR product).

Mouse toxicity studies

Mouse toxicity studies were performed using C57BL6 animals and were approved by the Fox Chase Cancer Center IACUC. Mice were injected intraperitoneally with the dose and frequency as indicated in figure legends, with the exception of Supplementary Fig. S3 (oral gavage) and Supplementary Fig. S5 (osmotic pump). For blood analysis, samples were taken by retro-orbital bleed and complete blood counts were performed using an Abaxis Vetscan5. At the end of experiments, mice were euthanized and liver, spleen, thymus, colon and bone marrow were collected, fixed in buffered formalin, and embedded in paraffin. The paraffin blocks were cut into 5-μm thick sections that were placed on positively charged slides. Sections were stained with hematoxylin and eosin (H&E) and mounted with Permount (Thermo Fisher Scientific). All H&E slides were viewed with a Nikon Eclipse 50i microscope and photomicrographs were taken with an attached Nikon DS-Fi1 camera.

For caspase-3 and Ki67 IHC staining, 5-μm formalin-fixed, paraffin-embedded sections were deparaffinized, hydrated, and subjected to heat-induced epitope retrieval with 0.01 mol/L citrate buffer (pH 6.0). Endogenous peroxidases were quenched by the immersion of slides in 3% hydrogen peroxide solution. Sections were incubated overnight with primary antibodies to anti-human ki-67 (1:100, catalog no. M7240, DAKO) and cleaved caspase-3 (1:200, catalog no. 9661 Cell Signaling Technology) at 4°C in a humidified slide chamber. Immunodetection was performed using the Dako Envision+ polymer system and immunostaining was visualized with the chromogen 3, 3′-diaminobenzidine. The sections were then washed, counterstained with hematoxylin, dehydrated with ethanol series, cleared in xylene, and mounted. As a negative control, the primary antibody was replaced with normal mouse/rabbit IgG to confirm absence of specific staining. Image analysis quantiation was performed by scanning with an Aperio ScanScope CS 5 slide scanner (Aperio). Scanned images were then viewed with Aperio's image viewer software (ImageScope, version 11.1.2.760, Aperio). Selected regions of interest were outlined manually by a pathologist (K.Q. Cai).

Mouse xenograft studies

C.B17 SCID mice were injected in the hind flank with 1 × 107 of tumor cells in 200 μL of Matrigel. Tumors were then allowed to grow until visible (1–4 weeks depending on line) and mice were randomly divided into vehicle and treatment groups. Mice were then treated with either vehicle or drug combination at a schedule indicated in figures. Tumor size was measured weekly by calipers as described in ref. 39 in all experiments except for Fig. 7A in which tumor volume was measured by MRI (40). Tumor growth was calculated by comparing the difference in volume at Tn with the volume at the time of the start of treatment (T0).

Statistical analysis

Pairwise comparisons between groups were performed using a Student two-sided t test. A P value of 0.05 or less was considered significant. For multiple group comparisons, single-sided ANOVA was used followed by post hoc test. All graphs show SEM.

MTA protects against 6TG toxicity in vitro, but not in vivo

We examined the effects of MTA addition on 6TG cell killing in a variety of MTAP+ and MTAP cell lines. For each line, we determined the IC50 of 6TG at 48 hours in the presence or absence of 10 μmol/L MTA and compared the ratio of the IC50 concentrations. In MTAP+ cells, we observed a 360-fold mean increase in IC50 concentration when MTA was added, but in MTAP cells, the mean increase was only 2-fold (Supplementary Fig. S1A; Supplementary Table S1). These findings confirm that MTA protects MTAP+ cells significantly better than MTAP cells from 6TG toxicity.

We next performed a series of in vivo experiments comparing the effects of 6TG alone versus 6TG+MTA. We injected (i.p.) C57BL6 mice with a single dose of either vehicle, 6TG (75 mg/kg), a combination of 6TG (75 mg/kg) + MTA (100 mg/kg), or a combination 6TG (75 mg/kg) + adenine (45 mg/kg). After four days, we examined toxicity by assessing white blood cell counts. As expected, mice treated with 6TG showed significantly reduced numbers white blood cells, lymphocytes, and neutrophils compared with untreated control animals (Supplementary Fig. S1B). However, we observed no beneficial effect of the addition of MTA or adenine. This result was surprising to us especially given the report by Bertino and colleagues that MTA could alleviate the lethality caused by multiple 6TG injections in SCID mice (38). Therefore, we performed an identical experiment in which we gave either SCID mice or C57BL6 mice, 6TG (75 mg/kg) or 6TG (75 mg/kg)+ MTA (100 mg/kg) using the same schedule reported in their article and followed the animals over time. In our hands, all of the animals in both groups died by 24 days from the start of treatment, and there was no difference between the non-MTA–treated and the MTA-treated animals (Supplementary Fig. S1C; Supplementary Fig. S2). A series of additional experiments were also performed in which we varied the mouse route of administration (oral and infusion pump), the timing of the MTA relative to the 6TG administration, and changed the ratio of 6TG to MTA (Supplementary Figs. S3–S6). None of the experiments showed any evidence that MTA could alleviate 6TG toxicity.

Finally, we performed a tissue toxicity study where the animals were injected with either vehicle, 6TG, or 6TG + MTA for four consecutive days. On the fifth day, the animals were euthanized and the intestine, spleen, bone marrow, and thymus were collected and analyzed for histopathology, and blood was collected for white blood cell analysis. Mice treated with 6TG exhibited the most severe toxicity in intestine and bone marrow (Supplementary Fig. S7A and S7B). The intestine showed severe mucosal atrophy, infiltration of inflammatory cells, and the basal portions of the glands were dilated. Bone marrow exhibited a dramatic decrease in cellularity. More modest atrophy was observed in the spleen and thymus. In blood, we observed a 40% decrease in white blood cells (P = 0.002), a 36% decrease in lymphocytes (P = 0.0017), and a 69% decrease in neutrophils (P = 0.0005; Supplementary Fig. S7C). Addition of MTA did not appear to significantly reverse toxicity in any of the tissues examined and actually made the neutropenia worse. Our experiments taken together show that MTA is unable to protect tissues against 6TG toxicity in vivo.

MTA protects against 2FA toxicity in an MTAP-dependent manner in cell culture

Our inability to ameliorate 6-TG toxicity by MTA or adenine in vivo was puzzling to us, especially given the robust protection we always observed in MTAP+ cells in vitro. One possible explanation is that in vivo PRPP is not rate limiting, so even in the presence of adenine, there is sufficient PRPP to facilitate the conversion to 6TG to 6TGMP. If this was the case, using a direct adenine analogue might bypass this problem as adenine and the adenine analogue would compete directly for binding to the APRT enzyme. To examine this possibility, we chose to examine 2FA, as fludarabine (2FA attached to a modified sugar–phosphate moiety) is used clinically to treat certain types of cancers (41). Interestingly, 2FA has never been tested in humans because early animal studies suggested it would be too toxic (42).

Our first series of experiments were cell line studies to determine whether MTA could protect against killing by 2FA in vitro. Initially, we examined the effects using an isogenic MTAP+ and MTAP cell line derived from a human HT1080 fibrosarcoma cell line previously developed in the laboratory (37). In MTAP cells, a 48-hour exposure to 2FA effectively killed all cells with a calculated IC50 of 75 nmol/L (Fig. 2A). The addition of 10 μmol/L MTA did not affect the IC50, whether added at the same time, 1, or 2 hours before 2FA exposure. In contrast, in MTAP+ cells, addition of MTA dramatically increased the IC50 from 27 nmol/L to 663 nmol/L. We also tested the effects of 2FA ± MTA on an isogenic set of MTAP+ and MTAP nontransformed mouse NIH 3T3 cells made by deleting MTAP using CRISPR. Again, we saw a dramatic shift in IC50 concentrations when MTA was added to 2FA only when MTAP was present (Fig. 2B). In addition to these isogenic cell lines, we also examined a number of other MTAP+ and MTAP cell lines for MTA protection. On average, we observed a 21-fold increase in the IC50 concentration in MTAP+ cells, but no significant change in MTAP cells (Fig. 2C; Supplementary Table S2). These studies suggest that MTA can protect against the effects of 2FA toxicity in cell lines similar to that observed for 6TG.

Figure 2.

Effect of MTA on 2FA IC50 in isogenic MTAP+ and MTAP cell lines. A, Dose–response curve for HT1080:MTAP and HT1080:MTAP+ cell lines treated in the absence or presence on 10 μmol/L MTA. Cells were given MTA either one or two hours before 2FA treatment, or at the same time. B, Dose–response curves for NIH3T3 mouse fibroblasts that have had MTAP deleted by CRISPR technology and nondeleted controls. C, Graph shows the ratio of calculated IC50 for 2FA either in the presence or absence of 10 μmol/L MTA for 13 different cell lines (Supplementary Table S2). *, P < 0.002. D, Mixing experiment. Top row shows PCR product using primers specific for wild-type MTAP, while bottom row shows PCR product specific toward MTAPΔ allele. DNA from untreated wild-type NIH3T3 cells (WT) or CRISPR MTAPΔ cells are shown in first two lanes (note some WT allele is still present in the CRISPR population). These two cell lines were then mixed in either a 1:1 or 4:1 ratio and treated with 1.25 μmol/L 2FA/10 μmol/L MTA for 70 hours or left untreated and PCR was performed on isolated DNA.

Figure 2.

Effect of MTA on 2FA IC50 in isogenic MTAP+ and MTAP cell lines. A, Dose–response curve for HT1080:MTAP and HT1080:MTAP+ cell lines treated in the absence or presence on 10 μmol/L MTA. Cells were given MTA either one or two hours before 2FA treatment, or at the same time. B, Dose–response curves for NIH3T3 mouse fibroblasts that have had MTAP deleted by CRISPR technology and nondeleted controls. C, Graph shows the ratio of calculated IC50 for 2FA either in the presence or absence of 10 μmol/L MTA for 13 different cell lines (Supplementary Table S2). *, P < 0.002. D, Mixing experiment. Top row shows PCR product using primers specific for wild-type MTAP, while bottom row shows PCR product specific toward MTAPΔ allele. DNA from untreated wild-type NIH3T3 cells (WT) or CRISPR MTAPΔ cells are shown in first two lanes (note some WT allele is still present in the CRISPR population). These two cell lines were then mixed in either a 1:1 or 4:1 ratio and treated with 1.25 μmol/L 2FA/10 μmol/L MTA for 70 hours or left untreated and PCR was performed on isolated DNA.

Close modal

We also performed a mixing experiment designed to determine whether MTAP+ cells could protect MTAP cells from 2FA+MTA toxicity in trans. We mixed wild-type and CRISPR MTAP-deleted NIH3T3 cells in either a 1:1 or 4:1 ratio and then treated them with either nothing or 2FA+MTA. After 70 hours, cells were harvested and allele-specific PCR was performed to determine the relative abundance of the MTAP+ and MTAP-deleted allele (Fig. 2D). After 2FA+MTA treatment, we were unable to detect the PCR product from the MTAP-deleted allele, but did see a strong band for the MTAP+ allele. These results indicate that MTAP+ cells do not protect MTAP cells from killing by 2FA+MTA in trans.

MTA protects against 2FA toxicity in mice

Some limited toxicology studies have been performed in mice for 2FA (see toxnet.nlm.nih.gov) showing that the LD50 for 2FA in mice (11.34 mg/kg) is significantly lower than the LD50 for 6TG (54 mg/kg). On the basis of this information, we performed an initial experiment in which C57BL6 mice were given 2FA at either 5 and 20 mg/kg in the presence or absence of MTA (100 mg/kg). Mice were given the drugs by intraperitoneal injection once a day for five days and then blood was analyzed eight days after the final injection (Fig. 3A). All of the animals survived to this time, and blood counts revealed that the 20 mg/kg 2FA dose caused a significant reduction in white blood cells and lymphocytes. Importantly, the mice treated with a combination of 2FA (20 mg/kg) + MTA (100 mg/kg) had lymphocyte counts indistinguishable from the control animals. On the basis of this result, we performed a follow-up study in which we obtained blood and weighed each animal both before and after treatment (Fig. 3B and C). In mice treated with 2FA alone lymphocyte counts decreased 70.2% (P < 0.0003) after treatment, but in animals treated with the 2FA+MTA combination, the decrease was only 21.4% (P < 0.04). In addition, while mice treated with 2FA alone lost on average 4.2 g of weight (P < 0.009), mice receiving 2FA+MTA only lost 0.3 g.

Figure 3.

Effects of 2FA and 2FA+MTA on mouse white blood cells (WBC). A, C57BL6 Mice were given 2FA or 2FA+MTA by intraperitoneal injection at the indicated dose for five consecutive days once a day for five days and then blood was analyzed eight days after the last injection. SEM is indicated. *, P < 0.05 compared with untreated control. B, Mouse blood was collected at baseline and then C57BL6 mice were given 2FA or 2FA+MTA by intraperitoneal injection for five consecutive days, followed by blood collection four days after last injection. SEM is indicated. *, P < 0.05 compared with baseline value. C, Change in weight for mice in experiment described in B.

Figure 3.

Effects of 2FA and 2FA+MTA on mouse white blood cells (WBC). A, C57BL6 Mice were given 2FA or 2FA+MTA by intraperitoneal injection at the indicated dose for five consecutive days once a day for five days and then blood was analyzed eight days after the last injection. SEM is indicated. *, P < 0.05 compared with untreated control. B, Mouse blood was collected at baseline and then C57BL6 mice were given 2FA or 2FA+MTA by intraperitoneal injection for five consecutive days, followed by blood collection four days after last injection. SEM is indicated. *, P < 0.05 compared with baseline value. C, Change in weight for mice in experiment described in B.

Close modal

We next examined the effects of 2FA (20 mg/kg) and 2FA (20 mg/kg) + MTA (100 mg/kg) on mouse tissues using a protocol identical to that used for 6TG. 2FA alone caused significant toxicity to all four tissues examined: small intestine, bone marrow, spleen, and thymus (Fig. 4A). Compared with 6TG, 2FA caused significantly more toxicity in small intestine, spleen, and thymus, but was somewhat less toxic with regards to bone marrow (Fig. 4B). In terms of blood cells, 2FA-treated mice exhibited greater lymphopenia, but did not exhibit neutropenia (Fig. 4C). Strikingly, addition of MTA to 2FA almost entirely prevented all the toxic effects of 2FA. Intestine, bone marrow, spleen, and thymus treated with the 2FA+MTA were all indistinguishable from untreated control samples. In this experiment, white blood cell and lymphocyte counts in 2FA+MTA were not statistically different from control animals. We also performed an additional study in which mice were injected with either 2FA (20 mg/kg), or 2FA (20 mg/kg) + MTA (100 mg/kg) for eight times over a 10-day period and then monitored for survival an additional 22 days. While all of the 2FA alone animals died by day 16, all of the 2FA+MTA animals were alive at day 28 (Supplementary Fig. S8). Together, these studies show that MTA can protect against 2FA related toxicity in vivo.

Figure 4.

2FA toxicity in mouse tissues. C57BL6 mice (n = 3/group) were injected with either vehicle, 20 mg/kg 2FA, or 20 mg/kg 2FA +100 mg/kg MTA for four consecutive days. Tissue was collected on day 5. A, Representative images of indicated H&E-stained tissues. B, H&E-stained images were rated blindly by a pathologist on a four point scale (1 = normal; 4 = severe cellular atrophy). Mean ratings for each tissue is shown. C, Bar chart showing various white blood cell measures in control and treated animals. Error bars, SEM. *, P < 0.05 compared with untreated control.

Figure 4.

2FA toxicity in mouse tissues. C57BL6 mice (n = 3/group) were injected with either vehicle, 20 mg/kg 2FA, or 20 mg/kg 2FA +100 mg/kg MTA for four consecutive days. Tissue was collected on day 5. A, Representative images of indicated H&E-stained tissues. B, H&E-stained images were rated blindly by a pathologist on a four point scale (1 = normal; 4 = severe cellular atrophy). Mean ratings for each tissue is shown. C, Bar chart showing various white blood cell measures in control and treated animals. Error bars, SEM. *, P < 0.05 compared with untreated control.

Close modal

Xenograft studies

We next examined the effectiveness of 2FA+MTA in inhibiting human MTAP tumor cell growth in SCID mice. For these studies, we used 10 mg/kg 2FA and 50 mg/kg MTA, as we found that SCID mice were somewhat less tolerant of the combination (Supplementary Fig. S9A and S9B). In our first study, we injected 1 × 107MTAP HT1080 cells subcutaneously and allowed them to form palpable tumors. Mice were then randomly divided and injected with either 2FA+MTA or vehicle alone (Fig. 5A). After 15 days, we observed a statistically significant decrease in tumor growth in the treated compared with the untreated mice. To show that tumor growth inhibition was dependent on loss of MTAP, we performed a second experiment in which we compared the response of isogenic MTAP and MTAP+ HT1080 cells to 2FA+MTA (Fig. 5B). We observed that in mice treated with 2FA+MTA, the relative growth rate of MTAP tumors was significantly slower than that observed for MTAP+ tumors. At the end of the experiment, tumors were excised and analyzed for MTAP activity, Ki67 staining, and caspase-3 staining. As expected, high levels of MTAP activity were detected only in the tumors formed by the MTAP+ cells (Fig. 5C). We found no difference in Ki67 staining, but saw a significant increase in the fraction of cell staining positive for caspase-3, suggesting that 2FA+MTA is stimulating apoptosis (Fig. 6).

Figure 5.

Effects of 2FA+MTA on HT1080 xenograft tumors. A, Comparison of tumor growth in SCID mice injected with MTAP HT1080 cells treated with vehicle and 10 mg/kg 2FA+ 50 mg/kg MTA (n = 8 per group). Arrows, days when injected. Error bars, SEM. *, P < 0.05 at indicated time. B, Comparison of growth rates in MTAP and MTAP+ HT1080 xenograft tumors treated with 10 mg/kg 2FA+ 50 mg/kg MTA (n = 5 per group). C, Quantification of MTAP activity of tumors isolated at the end of experiment B.

Figure 5.

Effects of 2FA+MTA on HT1080 xenograft tumors. A, Comparison of tumor growth in SCID mice injected with MTAP HT1080 cells treated with vehicle and 10 mg/kg 2FA+ 50 mg/kg MTA (n = 8 per group). Arrows, days when injected. Error bars, SEM. *, P < 0.05 at indicated time. B, Comparison of growth rates in MTAP and MTAP+ HT1080 xenograft tumors treated with 10 mg/kg 2FA+ 50 mg/kg MTA (n = 5 per group). C, Quantification of MTAP activity of tumors isolated at the end of experiment B.

Close modal
Figure 6.

Caspase-3 and Ki67 stainings of HT1080 MTAP+ and MTAP xenograft tumors treated with 2FA+MTA (n = 5/group). Left, representative images of staining for caspase-3 and KI67 from tumors on indicated genotype. Right, quantitation of slides as described in Materials and Methods.

Figure 6.

Caspase-3 and Ki67 stainings of HT1080 MTAP+ and MTAP xenograft tumors treated with 2FA+MTA (n = 5/group). Left, representative images of staining for caspase-3 and KI67 from tumors on indicated genotype. Right, quantitation of slides as described in Materials and Methods.

Close modal

We tested three additional MTAP tumor cell lines for growth inhibition by 2FA+MTA in SCID mice. A172 and U87 cells are glioblastoma-derived lines, while A549 cells are derived from a lung adenocarcinoma. Both tumor types show high rates of MTAP-loss. In mice containing U87 and A549 xenografts, we observed significant tumor growth inhibition in 2FA+MTA treated mice compared to vehicle treated mice (Fig. 7A and B). In mice bearing A172 xenograft tumors, treatment with 2FA+MTA resulted in a dramatic shrinkage of the tumors (Fig. 7C). These studies indicate that 2FA+MTA can inhibit the growth of a variety of MTAP-deleted tumor xenografts.

Figure 7.

Effect of 2FA+MTA combination on growth of MTAP U87, A172, and A549 tumor xenografts. A, Growth of vehicle-treated and 2FA+MTA–treated A549 tumors. Days of injection are shown by arrows. Error bars, SEM. *, P < 0.05 at indicated time. B, Growth of treated and untreated U87 tumors. C, Growth of treated and untreated A172 tumors. D, Weight of mice over time for U87 experiment.

Figure 7.

Effect of 2FA+MTA combination on growth of MTAP U87, A172, and A549 tumor xenografts. A, Growth of vehicle-treated and 2FA+MTA–treated A549 tumors. Days of injection are shown by arrows. Error bars, SEM. *, P < 0.05 at indicated time. B, Growth of treated and untreated U87 tumors. C, Growth of treated and untreated A172 tumors. D, Weight of mice over time for U87 experiment.

Close modal

In the U87 xenograft experiment, in which the animals received a total of 15 injections over a 36-day period, we also measured the weight of the animals (Fig. 7D). Mice initially lost about 15% of their body weight during the first week of treatment with 2FA+MTA, but after the first week, they started to gain weight at a rate that was similar to that observed in vehicle-treated animals. These data suggest that the 2FA+MTA combination can be given relatively frequently and for long duration without adverse effects.

In this article, we explore two different drug combinations, 6TG+MTA and 2FA+MTA, for their potential as agents to treat tumors that lack expression of MTAP. The underlying concept is that MTA will act as a “protecting agent” in cells that express MTAP, but will not protect in tumor cells lacking MTAP expression. This protection occurs because MTAP converts MTA to adenine, which is then converted to AMP by addition of a sugar and phosphate moiety that comes from PRPP. Because 6TG and 2FA both require conversion to either 6TGMP or 2FAMP to be toxic, it is hypothesized that the presence of high concentrations of adenine will reduce the PRPP pool, resulting in a reduction in the formation 6TGMP and 2FAMP. In cell culture, this mechanism is likely to be correct. Lubin found that addition of MTA to MTAP+ human fibroblasts cells protected these cells from the toxic effects of 2,6-diaminopurine, 6-methylpurine and 2FA, but addition of MTA did not protect MTAP tumor cells (36). Using a pair of genetically engineered isogenic MTAP+ and MTAP cell lines, our laboratory previously showed that addition of MTA shifted the IC50 of 6TG from 0.19 μmol/L to 3.7 μmol/L in MTAP+ cells, but did not shift the IC50 in MTAP cells (37). However, when adenine was added, both MTAP+ and MTAP cells were protected. In addition, treatment with a pharmacologic inhibitor of MTAP entirely abolished protection by MTA. In this article, we show that MTA can protect against 2FA toxicity to a similar extent.

The surprising finding from the mouse studies reported here was that while MTA protected against 2FA toxicity, it failed to protect against 6TG toxicity. We first focused on the 6TG+MTA combination because 6TG was already an FDA-approved drug and the two reports in the literature suggested that MTA could protect against 6TG toxicity (20, 38). Despite varying the amount of 6TG used, the ratio of MTA to 6TG, and the mode of delivery, we failed to see significant levels of protection by MTA. Toxicity in these studies was examined in a variety of assays including white blood cell counts, histopathology of various tissues, and survival. It should be noted that even when we used the identical concentrations of 6TG and MTA reported in the two previous reports, the exact same schedule, the same method of delivery, and the same outcome measure (survival), we were unable to repeat their observations. We have no explanation for this difference in results, although it is possible that there were subtle differences in the mouse strains that were used (C.B17 SCID vs. NOD SCID) or the timing of the administration of 6TG and MTA.

However, we did observe robust in vivo protection with the combination of 2FA and MTA. Treatment of C57BL6 mice with 20 mg/kg 2FA by itself resulted in severe toxicity in the small intestine, spleen, thymus, and moderate toxicity in the bone marrow. This pattern of toxicity was somewhat different from that observed with 75 mg/kg 6TG, which was particularly hard on the bone marrow, but not as severe on the other tissues. Amazingly, the addition 100 mg/kg MTA to 2FA almost completely eliminated the toxic effects. For the most part, tissues of mice treated with the 2FA+MTA combination were indistinguishable from control animals. Mice tolerated multiple injections quite well, and maintained weight while on treatment. An interesting question is: why did MTA protect against both 6TG and 2FA toxicity in vitro, but only 2FA in vivo? A possible explanation may be related to differences in what is rate-limiting in the mechanism of activation of the two drugs. In cell culture, it is likely that PRPP depletion by MTA is the critical component. Both 2FA and 6TG require PRPP for transformation into their respective nucleotide analogues and the absolute concentration of MTA added to the cell culture medium is high (10 μmol/L), thus PRPP depletion seems quite likely. However, in vivo it is unlikely that we are achieving these very high MTA levels, and even if we were, it is unknown if PRPP concentrations in tissues are similar to what is observed in cell culture. The fact that addition of adenine did not protect against 6TG toxicity (Fig. 2B) also strongly suggests that depletion of PRPP is not occurring in vivo. The most likely reason MTA protects against 2FA is that adenine and 2FA compete directly for binding to APRT and thus prevents the formation of 2FAMP. In contrast because 6TG uses GPRT for activation (43), there would be no direct completion between 6TG and adenine.

We used the combination of 2FA+MTA to examine growth inhibition of four different MTAP-deleted tumor cell lines in SCID mice as xenografts. In all cases, we found that treatment of mice with 2FA+MTA inhibited tumor growth compared with vehicle-treated controls. We also used isogenic HT1080 cell lines that were either MTAP+ or MTAP to examine the effect of MTAP status on treatment. As expected, we found that the 2FA+MTA combination was much more effective in inhibiting the growth of MTAP tumor cells than MTAP+ cells. We also observed a significantly higher percentage of cells from MTAP tumors that were positive for the apoptotic marker caspase-3, but observed no difference in the cell-cycle marker, Ki67. This finding may explain why, with one exception, we only observed a slowing of tumor growth and not regression of tumors in the mouse xenografts. We hypothesize that in fast-growing xenograft tumors the division rate is much faster than the apoptotic rate. Consistent with this idea, we observed regression in A172 cells, which also had the slowest growth rate.

While the results reported here are encouraging, there are many additional studies that will need to be performed before human clinical trials can begin. For all of the experiments reported here, we used a five to one ratio of MTA to 2FA and we have not explored other ratios. In addition, our use of 20 mg/kg 2FA and 100 mg/kg MTA was the maximum we could use given the solubility of these two compounds. Higher doses might be achievable by either optimizing solvent choice or by giving the compounds orally instead of by injection. It seems reasonable to assume that the higher the dose of 2FA that could be delivered, the more effective the treatment would be. Finally, pharmacodynamics studies will need to be performed to optimize drug dosing and delivery.

In summary, we show here that the combination of 2FA+MTA has potential as an agent to specifically target human tumors lacking expression of MTAP. Given the high frequency of MTAP loss in a variety of tumor types, we believe this combination is worth pursuing.

No potential conflicts of interest were disclosed.

Conception and design: B. Tang, H.-O. Lee, W.D. Kruger

Development of methodology: B. Tang, H.-O. Lee, K.Q. Cai

Acquisition of data (provided animals, acquired and managed patients, provided facilities, etc.): B. Tang, S.S. An, K.Q. Cai

Analysis and interpretation of data (e.g., statistical analysis, biostatistics, computational analysis): B. Tang, H.-O. Lee, K.Q. Cai, W.D. Kruger

Writing, review, and/or revision of the manuscript: H.-O. Lee, W.D. Kruger

Study supervision: W.D. Kruger

This work was supported by an NIH grant CA006927 (for all authors). We acknowledge the help from the Fox Chase Laboratory Animal, Biological Imaging, and Experimental Histopathology Facilities. We also acknowledge critical reviewing of the manuscript by Sapna Gupta. Some funding for these studies came from the Fox Chase Cancer Center core grant, NIH CA006927.

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.

1.
Toohey
JI
. 
Methylthioadenosine nucleoside phosphorylase deficiency in methylthio-dependent cancer cells
.
Biochem Biophys Res Commun
1978
;
83
:
27
35
.
2.
Carrera
CJ
,
Eddy
RL
,
Shows
TB
,
Carson
DA
. 
Assignment of the gene for methylthioadenosine phosphorylase to human chromosome 9 by mouse-human somatic cell hybridization
.
Proc Natl Acad Sci U S A
1984
;
81
:
2665
8
.
3.
Olopade
OI
,
Bohlander
SK
,
Pomykala
H
,
Maltepe
E
,
Van Melle
E
,
Le Beau
MM
, et al
Mapping of the shortest region of overlap of deletions of the short arm of chromosome 9 associated with human neoplasia
.
Genomics
1992
;
14
:
437
43
.
4.
Batova
A
,
Diccianni
MB
,
Nobori
T
,
Vu
T
,
Yu
J
,
Bridgeman
L
, et al
Frequent deletion in the methylthioadenosine phosphorylase gene in T- cell acute lymphoblastic leukemia: strategies for enzyme-targeted therapy
.
Blood
1996
;
88
:
3083
90
.
5.
Hori
Y
,
Hori
H
,
Yamada
Y
,
Carrera
CJ
,
Tomonaga
M
,
Kamihira
S
, et al
The methylthioadenosine phosphorylase gene is frequently co-deleted with the p16INK4a gene in acute type adult T-cell leukemia
.
Int J Cancer
1998
;
75
:
51
6
.
6.
Dreyling
MH
,
Roulston
D
,
Bohlander
SK
,
Vardiman
J
,
Olopade
OI
. 
Codeletion of CDKN2 and MTAP genes in a subset of non-Hodgkin's lymphoma may be associated with histologic transformation from low- grade to diffuse large-cell lymphoma
.
Genes Chromosomes Cancer
1998
;
22
:
72
8
.
7.
Bertino
JR
,
Lubin
M
,
Johnson-Farley
N
,
Chan
WC
,
Goodell
L
,
Bhagavathi
S
. 
Lack of expression of MTAP in uncommon T-cell lymphomas
.
Clin Lymphoma Myeloma Leuk
2012
;
12
:
306
9
.
8.
Usvasalo
A
,
Ninomiya
S
,
Raty
R
,
Hollmen
J
,
Saarinen-Pihkala
UM
,
Elonen
E
, et al
Focal 9p instability in hematologic neoplasias revealed by comparative genomic hybridization and single-nucleotide polymorphism microarray analyses
.
Genes Chromosomes Cancer
2010
;
49
:
309
18
.
9.
Schmid
M
,
Malicki
D
,
Nobori
T
,
Rosenbach
MD
,
Campbell
K
,
Carson
DA
, et al
Homozygous deletions of methylthioadenosine phosphorylase (MTAP) are more frequent than p16INK4A (CDKN2) homozygous deletions in primary non- small cell lung cancers (NSCLC)
.
Oncogene
1998
;
17
:
2669
75
.
10.
Watanabe
F
,
Takao
M
,
Inoue
K
,
Nishioka
J
,
Nobori
T
,
Shiraishi
T
, et al
Immunohistochemical diagnosis of methylthioadenosine phosphorylase (MTAP) deficiency in non-small cell lung carcinoma
.
Lung Cancer
2009
;
63
:
39
44
.
11.
Subhi
AL
,
Tang
B
,
Balsara
BR
,
Altomare
DA
,
Testa
JR
,
Cooper
HS
, et al
Loss of methylthioadenosine phosphorylase and elevated ornithine decarboxylase is common in pancreatic cancer
.
Clin Cancer Res
2004
;
10
:
7290
6
.
12.
Hustinx
SR
,
Hruban
RH
,
Leoni
LM
,
Iacobuzio-Donahue
C
,
Cameron
JL
,
Yeo
CJ
, et al
Homozygous deletion of the MTAP gene in invasive adenocarcinoma of the pancreas and in periampullary cancer: a potential new target for therapy
.
Cancer Biol Ther
2005
;
4
:
83
6
.
13.
Chen
YJ
,
Lin
SC
,
Kao
T
,
Chang
CS
,
Hong
PS
,
Shieh
TM
, et al
Genome-wide profiling of oral squamous cell carcinoma
.
J Pathol
2004
;
204
:
326
32
.
14.
Karikari
CA
,
Mullendore
M
,
Eshleman
JR
,
Argani
P
,
Leoni
LM
,
Chattopadhyay
S
, et al
Homozygous deletions of methylthioadenosine phosphorylase in human biliary tract cancers
.
Mol Cancer Ther
2005
;
4
:
1860
6
.
15.
Brat
DJ
,
James
CD
,
Jedlicka
AE
,
Connolly
DC
,
Chang
E
,
Castellani
RJ
, et al
Molecular genetic alterations in radiation-induced astrocytomas
.
Am J Pathol
1999
;
154
:
1431
8
.
16.
Suzuki
T
,
Maruno
M
,
Wada
K
,
Kagawa
N
,
Fujimoto
Y
,
Hashimoto
N
, et al
Genetic analysis of human glioblastomas using a genomic microarray system
.
Brain Tumor Pathol
2004
;
21
:
27
34
.
17.
Crespo
I
,
Tao
H
,
Nieto
AB
,
Rebelo
O
,
Domingues
P
,
Vital
AL
, et al
Amplified and homozygously deleted genes in glioblastoma: impact on gene expression levels
.
PLoS One
2012
;
7
:
e46088
.
18.
Garcia-Castellano
JM
,
Villanueva
A
,
Healey
JH
,
Sowers
R
,
Cordon-Cardo
C
,
Huvos
A
, et al
Methylthioadenosine phosphorylase gene deletions are common in osteosarcoma
.
Clin Cancer Res
2002
;
8
:
782
7
.
19.
Curtis
C
,
Shah
SP
,
Chin
SF
,
Turashvili
G
,
Rueda
OM
,
Dunning
MJ
, et al
The genomic and transcriptomic architecture of 2,000 breast tumours reveals novel subgroups
.
Nature
2012
;
486
:
346
52
.
20.
Collins
CC
,
Volik
SV
,
Lapuk
AV
,
Wang
Y
,
Gout
PW
,
Wu
C
, et al
Next generation sequencing of prostate cancer from a patient identifies a deficiency of methylthioadenosine phosphorylase, an exploitable tumor target
.
Mol Cancer Ther
2012
;
11
:
775
83
.
21.
Stadler
WM
,
Olopade
OI
. 
The 9p21 region in bladder cancer cell lines: large homozygous deletion inactivate the CDKN2, CDKN2B and MTAP genes
.
Urol Res
1996
;
24
:
239
44
.
22.
Illei
PB
,
Rusch
VW
,
Zakowski
MF
,
Ladanyi
M
. 
Homozygous deletion of CDKN2A and codeletion of the methylthioadenosine phosphorylase gene in the majority of pleural mesotheliomas
.
Clin Cancer Res
2003
;
9
:
2108
13
.
23.
Zimling
ZG
,
Jorgensen
A
,
Santoni-Rugiu
E
. 
The diagnostic value of immunohistochemically detected methylthioadenosine phosphorylase deficiency in malignant pleural mesotheliomas
.
Histopathology
2012
;
60
:
E96
105
.
24.
Wong
YF
,
Chung
TK
,
Cheung
TH
,
Nobori
T
,
Chang
AM
. 
MTAP gene deletion in endometrial cancer
.
Gynecol Obstet Invest
1998
;
45
:
272
6
.
25.
Meyer
S
,
Fuchs
TJ
,
Bosserhoff
AK
,
Hofstadter
F
,
Pauer
A
,
Roth
V
, et al
A seven-marker signature and clinical outcome in malignant melanoma: a large-scale tissue-microarray study with two independent patient cohorts
.
PLoS One
2012
;
7
:
e38222
.
26.
Powell
EL
,
Leoni
LM
,
Canto
MI
,
Forastiere
AA
,
Iocobuzio-Donahue
CA
,
Wang
JS
, et al
Concordant loss of MTAP and p16/CDKN2A expression in gastroesophageal carcinogenesis: evidence of homozygous deletion in esophageal noninvasive precursor lesions and therapeutic implications
.
Am J Surg Pathol
2005
;
29
:
1497
504
.
27.
Kim
J
,
Kim
MA
,
Min
SY
,
Jee
CD
,
Lee
HE
,
Kim
WH
. 
Downregulation of methylthioadenosin phosphorylase by homozygous deletion in gastric carcinoma
.
Genes Chromosomes Cancer
2011
;
50
:
421
33
.
28.
Huang
HY
,
Li
SH
,
Yu
SC
,
Chou
FF
,
Tzeng
CC
,
Hu
TH
, et al
Homozygous deletion of MTAP gene as a poor prognosticator in gastrointestinal stromal tumors
.
Clin Cancer Res
2009
;
15
:
6963
72
.
29.
Andor
N
,
Graham
TA
,
Jansen
M
,
Xia
LC
,
Aktipis
CA
,
Petritsch
C
, et al
Pan-cancer analysis of the extent and consequences of intratumor heterogeneity
.
Nat Med
2016
;
22
:
105
13
.
30.
Krasinskas
AM
,
Bartlett
DL
,
Cieply
K
,
Dacic
S
. 
CDKN2A and MTAP deletions in peritoneal mesotheliomas are correlated with loss of p16 protein expression and poor survival
.
Mod Pathol
2010
;
23
:
531
8
.
31.
He
HL
,
Lee
YE
,
Shiue
YL
,
Lee
SW
,
Chen
TJ
,
Li
CF
. 
Characterization and prognostic significance of methylthioadenosine phosphorylase deficiency in nasopharyngeal carcinoma
.
Medicine
2015
;
94
:
e2271
.
32.
Kryukov
GV
,
Wilson
FH
,
Ruth
JR
,
Paulk
J
,
Tsherniak
A
,
Marlow
SE
, et al
MTAP deletion confers enhanced dependency on the PRMT5 arginine methyltransferase in cancer cells
.
Science
2016
;
351
:
1214
8
.
33.
Marjon
K
,
Cameron
MJ
,
Quang
P
,
Clasquin
MF
,
Mandley
E
,
Kunii
K
, et al
MTAP deletions in cancer create vulnerability to targeting of the MAT2A/PRMT5/RIOK1 axis
.
Cell Rep
2016
;
15
:
574
87
.
34.
Mavrakis
KJ
,
McDonald
ER
 III
,
Schlabach
MR
,
Billy
E
,
Hoffman
GR
,
deWeck
A
, et al
Disordered methionine metabolism in MTAP/CDKN2A-deleted cancers leads to dependence on PRMT5
.
Science
2016
;
351
:
1208
13
.
35.
Robak
P
,
Robak
T
. 
Older and new purine nucleoside analogs for patients with acute leukemias
.
Cancer Treat Rev
2013
;
39
:
851
61
.
36.
Lubin
M
,
Lubin
A
. 
Selective killing of tumors deficient in methylthioadenosine phosphorylase: a novel strategy
.
PLoS One
2009
;
4
:
e5735
.
37.
Tang
B
,
Testa
JR
,
Kruger
WD
. 
Increasing the therapeutic index of 5-fluorouracil and 6-thioguanine by targeting loss of MTAP in tumor cells
.
Cancer Biol Ther
2012
;
13
:
1082
90
.
38.
Bertino
JR
,
Waud
WR
,
Parker
WB
,
Lubin
M
. 
Targeting tumors that lack methylthioadenosine phosphorylase (MTAP) activity: current strategies
.
Cancer Biol Ther
2011
;
11
:
627
32
.
39.
Jensen
MM
,
Jorgensen
JT
,
Binderup
T
,
Kjaer
A
. 
Tumor volume in subcutaneous mouse xenografts measured by microCT is more accurate and reproducible than determined by 18F-FDG-microPET or external caliper
.
BMC Med Imaging
2008
;
8
:
16
.
40.
Hensley
H
,
Quinn
BA
,
Wolf
RL
,
Litwin
SL
,
Mabuchi
S
,
Williams
SJ
, et al
Magnetic resonance imaging for detection and determination of tumor volume in a genetically engineered mouse model of ovarian cancer
.
Cancer Biol Ther
2007
;
6
:
1717
25
.
41.
Lukenbill
J
,
Kalaycio
M
. 
Fludarabine: a review of the clear benefits and potential harms
.
Leuk Res
2013
;
37
:
986
94
.
42.
Skipper
HE
,
Montgomery
JA
,
Thomson
JR
,
Schabel
FM
 Jr
. 
Structure-activity relationships and cross-resistance observed on evaluation of a series of purine analogs against experimental neoplasms
.
Cancer Res
1959
;
19
:
425
37
.
43.
Hill
DL
,
Pittillo
RF
. 
Use of Escherichia coli mutants to evaluate purines, purine nucleosides, and analogues
.
Antimicrob Agents Chemother
1973
;
4
:
125
32
.