The pyrimidine trifluorothymidine (TFT) inhibits thymidylate synthase (TS) and can be incorporated into the DNA. TFT, as part of TAS-102, is clinically evaluated in phase II studies as an oral chemotherapeutic agent. Erlotinib is a tyrosine kinase inhibitor of the epidermal growth factor receptor (EGFR) that is often deregulated in colorectal cancer. This study investigated molecular mechanisms underlying the cytotoxic actions of the combination of an EGFR−tyrosine kinase inhibitor with TFT in colorectal cancer cells Caco2, WiDR, Lovo92, and Colo320. Drug interactions were examined by the sulforhodamine B assay and subsequent combination index (CI) analyses, cell cycle effects by FACS analysis of propidium iodide stained cells, Akt, MAPK and EGFR phosphorylation and expression levels by Western blotting and TS activity by the TS in situ assay. All combination schedules were synergistic in wt-EGFR expressing (but with mutated downstream pathways) WiDR and Lovo92 (CI 0.4–0.8) and very synergistic in Caco2 cells (with wt-EGFR and functional downstream pathways; CI 0.1–0.3), but in EGFR-lacking Colo320 cells, no additional activity was found (CI 1.0–1.2). Synergism was mostly related to the induction of cell cycle arrest and an erlotinib-mediated inhibition of the pro-survival signaling through Akt and MAPK that was activated (phosphorylated) by TFT. Erlotinib inhibited TS activity in EGFR-expressing cell lines, probably due to cell cycle arrest in the G1 phase. TS activity was slightly lower in the combinations, probably due to cell cycle interference. Taken together, the combination of erlotinib with TFT seems to present a potential strategy in the field of molecular therapeutics. (Cancer Sci 2009)
Colorectal cancer is the third leading cause of cancer death worldwide. The standard therapy consists of 5FU and leucovorin, which is usually combined with either the topoisomerase I inhibitor irinotecan or the novel platinum analog oxaliplatin.(1) Addition of the latter compounds substantially increased the median survival. Further improvement of therapy can be expected from drugs bypassing 5FU resistance. Therefore 5FU prodrugs are currently being investigated, including S-1, UFT and capecitabine.(2) Recently, we and others showed that another fluoropyrimidine, TFT, might be more effective in CRC cells to overcome (acquired) 5FU and/or 5-fluoro-2′-deoxyuridine (FdUrd) resistance.(3,4) TFT is part of the oral fluoropyrimidine formulation TAS-102 and can be activated to its phosphate derivatives. When monophosphorylated, TFT can inhibit TS, thereby inhibiting the synthesis of new pyrimidines.(3) In its triphosphate form, TFT can be incorporated into the DNA, causing DNA damage. Both events will lead to the induction of cell death. TAS-102 is currently being tested in a phase II clinical trial against colon and gastric cancer.(3) TAS-102 is a promising novel formulation that also showed pronounced synergism with irinotecan and oxaliplatin.(5,6)
Epidermal growth factor receptor (EGFR) is overexpressed or deregulated in many human cancer types, including CRC.(7) A high EGFR expression level has been related to a poor prognosis.(8,9) Targeting EGFR might be a rational approach to treat CRC patients. EGFR inhibitors are under investigation, including the EGFR-TKIs gefitinib and erlotinib. Erlotinib is a highly potent reversible inhibitor of the tyrosine kinase domain of EGFR. It is an oral compound that is active against a wide range of colon cancers in vitro,(10) but its clinical application is mostly limited to NSCLC in which it is active against patients with activating EGFR mutations and wild-type k-Ras. Erlotinib as a monotherapy has also shown activity in metastatic CRC patients,(11) providing a basis for further studies in phase II in combination with chemotherapy. However, a proper analysis of the mechanism underlying the combination is essential. Although a combination of EGFR inhibition with chemotherapeutic regimens may enhance the anticancer response,(12,13) the interaction might be dependent on schedule and concentration. The combination of the multitargeted TS inhibitor pemetrexed with erlotinib showed a sequence-dependent synergistic interaction, for example, when cells were pre-exposed to pemetrexed in NSCLC cell lines.(14,15) Various combination studies with erlotinib are currently under investigation in advanced and metastatic CRC, including with capecitabine and oxaliplatin, capecitabine and irinotecan, but also with bevacizumab and 5FU, leucovorin, and oxaliplatin.(16–18)
The aim of our study was to investigate whether inhibition of EGFR could enhance the effect of TFT against CRC cells in vitro. We used erlotinib as a model TKI and investigated schedule dependency in relation to the phosphorylation levels of EGFR as well as that of the intracellular kinases Akt and MAPK. Furthermore, we assessed the effect on cell cycle distribution, modulation of TS activity, and the induction of DNA damage to gain insight into the mechanism underlying the synergistic effect of combination therapy with TFT and an EGFR-TKI.
Materials and Methods
Cell lines and chemicals Human colon carcinoma cell lines WiDR, Colo320, Lovo92 and Caco2 were cultured as monolayers in DMEM supplemented with 10% heat-inactivated FCS and 20 mM HEPES. Cells were maintained in a humidified 5% CO2 atmosphere at 37°C. TFT was provided by Taiho Pharmaceutical Co. (Tokushima, Japan).
Drug cytotoxicity assays Drug cytotoxicity was determined by the SRB assay as described previously.(19) In 96-well plates (Greiner Bio-One, Frickenhausen, Germany), 5000 cells/well were seeded in 100 μL medium. After 24 h, 100 μL drug containing medium was added to the wells. After 72 h drug exposure, cells were precipitated for 1 h at 4°C with 25 μL 50% trichloroacetic acid and colored with SRB (0.4% SRB in 1% acetic acid [w/v]). The optical density was measured at 540 nm after reconstitution of the dye in 150 μL 10 mM Tris. The IC50 values were estimated from graphs and are given in means ± SEM. For combination experiments, a concentration range which was fitted according to CalcuSyn (Biosoft, Cambridge, UK) was used with a fixed ratio based on the IC50. Cells were either exposed concurrently to both TFT and erlotinib for 72 h or 24 h pre-incubated with either TFT or erlotinib, followed by a 48 h exposure to the combination. The interaction was determined with the multiple drug effect method, in which a CI was calculated with CalcuSyn as described previously.(5) This method is distinct from other methods by the fact that both “potency” and “shape” of dose–effect curves of drugs and their combinations are taken into account. For calculation of the CI, only values above a fraction affected of 0.5 were used, equivalent to 50–100% growth inhibition. Per experiment, CI values at fraction affected 0.5, 0.75, and 0.9 were averaged according to the guidelines(5) and the mean was used for comparison of separate experiments. A CI <0.9 indicates synergism and >1.1 antagonism.
FACS analysis of cell cycle Cell cycle analysis and cell death measurement were carried out as described previously.(19) The time-course of cell cycle changes for various combinations with TFT was determined earlier, showing the most pronounced changes after 72 h. Therefore, we focused on 72 h treatment with IC50 concentrations of the drugs and drug combinations; this also enabled the cells to have completed enough cell cycles (at least three cycles) to determine statistically significant differences. In brief, 100 000 cells were seeded in six-well plates. Cells were trypsinized, resuspended in medium collected from the matching sample, and centrifuged for 4 min at 1000g. Subsequently, cells were stained with PI (0.1 mg/mL) in the dark on ice. DNA content of the cells was analyzed by FACS (FACScan; BD Immunocytometry Systems, San Jose, CA, USA) with an acquisition of 10 000 events. Sub-G1 peak was used as the indication of cell death induction.
Thymidylate Synthase in situ Activity
Inhibition of TS in intact cells was determined by measuring the conversion of [3H]-dCyd to 3H2O, which is catalyzed by TS as described previously.(20) Briefly, 1.106 cells were seeded and incubated with the IC50 concentrations of the different drugs and the simultaneous combination for 22 h. Subsequently, [3H]-dCyd (final concentration, 1 μM; specific activity 4.9 Ci/mmol) was added to each sample for 2 h at 37°C. Blanks consisted of culture medium only and untreated cells were used as controls. The reaction was stopped by adding trichloroacetic acid and unconverted [3H]-dCyd was removed by activated charcoal. After centrifugation, the supernatant was transferred to a liquid scintillation vial and counted.
Western blot analysis Cells were seeded in T25 culture flasks at a density of 1.5.106 cells. After 24 h, cells were exposed to IC50 concentrations of TFT or the various combination schedules (IC50 concentration) as described in the drugs cytotoxicity assay. After treatment for the indicated time-points in the figures, cells were lysed in lysis buffer (10% glycerol, 5 mM EDTA, 10 mM Tris-HCl [pH 7.5], 150 mM NaCl, 50 mM β-glycerophophate, 1% Triton X-100, 0.04% protease inhibitor cocktail, 0.1% 1 M NaVO3) and centrifuged at 11 000g at 4°C for 10 min. Protein concentration in the supernatant was determined by carrying out a Bio-Rad protein assay according to the manufacturer’s instructions (Bio-Rad Laboratories, Veenendaal, the Netherlands). From each condition 30 μg of protein was separated on a 10–12% SDS–PAGE and electroblotted onto a PVDF membrane. Blots were blocked in 5% milk in TBST (0.15 M NaCl, 0.05% Tween-20, 10 mM Tris–HCl [pH 8.0]) and subsequently incubated at 4°C overnight with the following antibodies: EGFR (#2232); phospho-EGFR (Tyr1068 #2234); Akt (#9272); phospho-Akt (Ser473 #9271); p42/44 MAPK (#9102); phospho-p42/44 MAPK (Thr202/Tyr204 #9101) (1 : 1000; Cell Signaling Technology, Danvers, MA, USA); p53 (AB-2; Oncogene Research Products, Cambridge, MA, USA); p21 (Santa Cruz Biotechnology, Santa Cruz, CA, USA); or phospho-γ-H2Ax (Ser139 #05-636) (1 : 1000; Upstate, Billerica, MA, USA) and β-Actin (#A5441) (1:10,000; Sigma Aldrich Chemicals, Deisenhofer, Germany) served as the loading control for protein amount. After the first antibody, secondary antibody was added and the signal was detected using ECL or ECL-plus on hyperfilms (Amersham International, Chalfont St Giles, UK).
Growth inhibition and combination analysis WiDR, Colo320, and Lovo92 cells showed comparable levels of sensitivity to TFT treatment as determined by the SRB assay (Table 1; Fig. 1). Cells were slightly more sensitive to TFT than to erlotinib. There was barely any difference in sensitivity to erlotinib in the three cell lines, which were all relatively insensitive with each IC50 in the μM range. This is probably because they have a different mutation status for several kinases, including EGFR (Table 1). Furthermore, they have different expression levels of p53 and p21 (Fig. 2). In WiDR cells, all combinations were synergistic. All combinations in Lovo92 cells were moderately synergistic, although at lower concentration ranges (growth inhibition <50%) the combinations were not synergistic (Fig. 1). However, we do not consider this low concentration range of therapeutic interest. In Colo320 cells the tested combinations were not synergistic, although antagonism was not observed. In contrast to previous experiences with erlotinib combined with a TS inhibitor,(14,15) the order of administration of the combination schedule was not important to achieve synergism.
Table 1. Characteristics of colorectal cancer cell lines used in this analysis
IC50 erlotinib (μM)
IC50 TFT (μM)
†Colo320 is w.t. p53, but lacks p21 (see Fig. 1) and therefore behaves as a mutant. EGFR, epidermal growth factor receptor; mut, mutant; TFT, trifluorothymidine; w.t., wild type.
8.2 ± 0.8
3.2 ± 0.3
4.4 ± 1.3
0.9 ± 0.1
3.2 ± 0.3
0.6 ± 0.1
0.3 ± 0.1
25 ± 2.0
In order to determine whether the sensitivity to erlotinib is important for synergism with TFT, we included Caco2, a cell line that is very sensitive to erlotinib (Fig. 1) due to a wild-type EGFR expression and with no reported mutations in the genes of p53, Braf, k-Ras, or PTEN. In addition, EGFR expression levels were higher in Caco2 than in WiDR cells.(21) Caco2 cells were sensitive to erlotinib with an IC50 value in the nM range (Table 1; Fig. 1). In these cells, all three combination schedules were highly synergistic (Fig. 1).
Effects on cell cycle distribution To determine whether the synergistic actions were related to specific cell cycle effects, FACS analysis of PI stained cells was carried out using IC50 concentrations (Table 1; Fig. 3; Fig. S1, Supporting Information). TFT alone induced predominantly a G2/M-phase arrest, although this was cell line- and time-dependent. In WiDR and Lovo92, erlotinib alone increased cells in the G1-phase, which was not time-dependent. In contrast, in Colo320 cells, erlotinib caused no significant changes in the cell cycle distribution. The simultaneous exposure increased WiDR cells in the G1 phase to some extent. When preincubated with erlotinib, the combined treatment resulted in an increased S phase fraction. In Lovo92 cells, all drug combinations induced G1 phase arrest. In addition, in both WiDR and Lovo92 cells, the cell cycle distribution seen after combined drug treatments is more comparable to that of erlotinib alone than to that of TFT alone. When Colo320 cells were preincubated with erlotinib prior to addition of the combination, cells mostly accumulated in the S phase, whereas the other two combinations arrested cells mainly in the G2/M phase, comparable to TFT alone. Because of the different cell cycle distribution in the pre-erlotinib schedule, erlotinib may have off-target effects. In Colo320 cells the combinations were not synergistic. These different effects on cell cycle distribution between the various cell lines indicate that the interactions between the drugs are cell cycle mediated.
Effects on cell death induction To determine whether the combinations resulted in an increase in cell death induction, we analyzed the sub-G1-fraction of PI stained cells. In order to determine apoptosis we also determined caspase activation, which might not be seen in the sub-G1. The pattern of sub-G1 accumulation agreed with caspase activation (data not shown). In all three cell lines, TFT induced cell death in a time-dependent manner (Fig. 4). Erlotinib did not induce cell death directly. In WiDR, cell death was induced more strongly by the combinations compared to the control, although the combinations did not have a higher cell death than induced by 72 h TFT alone. In Lovo92 and Colo320 cells, cell death induced by the combination where TFT was given first was significantly higher than control levels, although lower than induced by TFT alone. As the combinations in EGFR expressing cells were synergistic, this may indicate that the combinations act by the induction of cell cycle arrest rather than cell death.
Thymidylate synthase activity Thymidylate synthase is an important cell cycle enzyme that plays a limiting role in de novo pyrimidine deoxynucleotide synthesis. As the cell cycle effects seem to be important for the synergistic action of the combinations, and TS is one of the targets of TFT, we determined TS inhibition in intact cells treated with TFT and erlotinib (Table 2). Based on previous time-course experiments with TS inhibitors in these cell lines, we chose 24 h to measure the inhibition.(22) TFT markedly inhibited TS activity in all three cell lines. Erlotinib alone also inhibited TS activity in EGFR expressing WiDR and Lovo92 cells. After a simultaneous combination, TS was inhibited to a larger extent than by TFT alone. This level of inhibition was almost similar to the expected level of inhibition, although the decrease was not significant (Table 2). The effect of erlotinib alone on TS might be related to a cell cycle-dependent activity of TS.(15,23)
Table 2. Thymidylate synthase in situ activity in WiDR, Lovo92 and Colo320 colorectal cancer cells
Thymidylate synthase in situ activity after 24 h exposure to IC50 concentrations of either trifluorothymidine (TFT), erlotinib, or the simultaneous combination. Values represent the percentage of untreated control cells and are expressed as means of at least three independent experiments ± SEM. Expected values are calculated by multiplying the percentage inhibition by erlotinib by the percentage activity of TFT alone. When the value is lower than the expected value, this means that thymidylate synthase inhibition is synergistic.
2.1 ± 0.6
37.4 ± 4.8
0.8 ± 0.2
33.6 ± 2.2
64.3 ± 10.2
24.3 ± 4.8
20.4 ± 2.2
101.5 ± 7.2
13.2 ± 3.4
Effects on EGFR, MAPK and Akt phosphorylation Erlotinib inhibits EGFR phosphorylation and thereby its downstream targets, Akt and MAPK. To determine whether cell cycle arrest was related to inhibition of these important cell growth signalling pathways, Western blot analysis was carried out. Lovo92 strongly expressed EGFR, WiDR had a moderate EGFR expression, but Colo320 did not express EGFR (Fig. 2).
As effects on phosphorylation are direct, the phosphorylation status of these proteins was determined after 2 h exposure of the drugs and simultaneous drug combination (Fig. 5). Moreover, to determine whether erlotinib can effectively inhibit EGFR, EGF was added to the cultures 5 min prior to cell lysis. In this way, it can be determined whether the EGFR signalling pathway was inhibited. In WiDR and Lovo92 cells, EGFR, Akt, and MAPK phosphorylation increased after EGF stimulation. TFT slightly increased the phosphorylation levels of MAPK. Erlotinib alone increased EGFR phosphorylation, but prevented the stimulation by EGF. Erlotinib clearly prevented activation of the downstream kinase Akt and to a lesser extent that of MAPK. In the combination, phosphorylation of EGFR and Akt were inhibited almost completely, whereas the stimulation of MAPK by EGF and TFT was completely prevented. In Colo320 cells, EGFR, Akt, and MAPK phosphorylation levels did not change after stimulation with EGF. EGFR was not detectable and could not be upregulated to a detectable level by any of the tested drugs or drug combinations. Akt phosphorylation was slightly decreased after TFT alone. These results indicate that erlotinib can inhibit EGFR signalling in cells with an active EGFR, but has no effect on downstream signaling in cells with low or absent EGFR expression.
In order to investigate long-term effects, and the effects in the sequential combinations, which can not be determined after 2 h and in which EGF addition does not play a role (data not shown), we also investigated the effects of the combination after 72 h (Fig. 6). Surprisingly, in WiDR and Lovo92 cells, after 48 h of TFT exposure phosphorylated EGFR increased, which was reduced after 72 h. In both WiDR and Lovo92 cells TFT increased phosphorylation levels of MAPK and Akt time-dependently. Erlotinib barely affected Akt phosphorylation, but in WiDR cells, phosphorylated Akt increased after 24 h, which reduced in time.
In WiDR and Lovo92, addition of erlotinib prevented the pro-survival signalling that was induced by TFT, even to lower levels than the control phosphorylation levels. Although after the sequential combination where TFT was given first, Akt and MAPK were still activated, this was reduced to control levels. In Colo320 cells, which did not express detectable EGFR or phosphorylated MAPK, no changes in expression and/or phosphorylation levels were observed after all different drugs and drug combinations. Overall, the inhibition of cell growth can be related to cell cycle effects, which in turn are related to the inhibition of the pro-survival signals that are induced by TFT.
Effect on DNA damage Recent reports have indicated that EGFR inhibition might decrease DNA repair activity,(24,25) therefore we determined the level of DNA damage after 72 h of drug incubation and whether DNA damage levels decreased (due to repair) after 48 h of growth in drug-free medium (Fig. 7). DNA damage was monitored by determination of the phosphorylation status of histone γ–H2Ax, one of the first events in DNA damage response. Cells exposed to the drugs or drug combinations induced γ-H2Ax phosphorylation, although for the combinations a less than additive effect was seen. In Lovo92 and WiDR cells, this DNA damage increased significantly (up to 55-fold) following subsequent 48 h of culture in drug-free medium. In Colo320 cells this accumulation did not increase, but may explain the observed cell cycle changes after exposure to the combination.
This study shows that a combination of TFT with erlotinib is synergistic in EGFR expressing colon cancer cells, also in the presence of a k-Ras mutation. This sensitization of moderately sensitive cells to erlotinib indicates the potential for evaluating the combination of TFT with EGFR-TKIs. Ongoing clinical studies are confirming the promising therapeutic activity of second-generation EGFR-TKIs and multitargeted TKIs in several tumor types.(26,27) The combination of TFT with erlotinib was strongly synergistic in Caco-2 cells with a functional (e.g. wild-type) EGFR signaling pathway, and the combination was synergistic in WiDR and Lovo92 cells that were moderately sensitive to erlotinib with mutations in the downstream EGFR pathway. The combination was not synergistic in cells that did not express EGFR. This underlines the importance of an active EGFR signaling pathway in the synergistic interaction. The mechanism underlying the synergistic interaction was probably due to inhibition of the downstream EGFR pro-survival signaling pathway and the induction of DNA damage.
Synergistic actions with EGFR-TKIs have previously been reported with cytotoxic agents such as 5FU and TS inhibitors (pemetrexed) and irinotecan, and with radiation.(15,28–30) These cytotoxic agents and radiation could increase the phosphorylation level of EGFR, which possibly reflects the activation of pro-survival signaling. This effect was blocked by addition of EGFR-TKIs, explaining the synergistic activities. 5FU or pemetrexed-induced activation of EGFR phosphorylation was prevented by gefitinib and erlotinib.(15,31) In our study, such a synergistic interaction was only found for cells that constitutively expressed both EGFR and activated the pro-survival signaling after exposure to TFT. TFT alone activated EGFR phosphorylation levels, and highly increased the phosphorylation levels of the pro-survival kinases MAPK and Akt. This pro-survival signaling could be inhibited by the addition of erlotinib, possibly explaining the synergistic interaction. In EGFR lacking Colo320 cells, erlotinib could inhibit cell growth and no synergism was found. In these cells TFT hardly activated pro-survival signaling and erlotinib did not have any effect on the phosphorylation levels of the tested intracellular kinases or on the cell cycle. Therefore, it is likely that the mechanism of erlotinib in these cells do not involve changes in intracellular signaling in the examined routes.
All three cell lines were moderately sensitive to erlotinib. In NSCLC and CRC, mutations in k-Ras and BRAF are potential biomarkers for erlotinib sensitivity. Mutations in these genes can cause constitutive activation of MAPK. The low EGFR expression in one cell line and k-Ras mutations may explain the low sensitivity to erlotinib. However, independent of the mutations in k-Ras, downstream signaling of Akt and MAPK was significantly inhibited in our cell lines. In addition, EGFR status is known not to be a predictive factor for responses to cetuximab, as responses were reported in patients with EGFR-negative tumors.(32,33) In addition, erlotinib has shown to have some off-target effects in leukemia cells, in which the JAK2/STAT5 pathway was inhibited,(34) and which were likely to be comparable to the off-target effects of gefitinib.(35) In the latter studies, differentiation and cell cycle blockade induced by EGFR antagonists exerting off-target effects on Acute Myeloid Leukemia (AML) cells were not automatically linked to an apoptotic response.(34,35) Our data indicate that erlotinib also has off-target effects in colon cancer cells. Ongoing clinical studies, in which the responses to the new-generation EGFR-TKIs are evaluated, appear to be promising. In this respect, based on our current data, the inhibition of pro-survival signaling by EGFR-TKIs can enhance the efficacy of the chemotherapeutic agent and may be worthwhile to test in a clinical setting.
Recent studies reported that ERGF-TKIs can decrease the expression levels and activity of TS in NSCLC, breast, and colorectal cancer cells(15,23,36) and could be related to synergism. As erlotinib did not inhibit TS itself in cell-free extracts,(15) the decrease in TS can be explained by the G1/G0 cell cycle arrest that was induced by the EGFR-TKI. TS is an enzyme that is only active in the S phase of the cell cycle. TS activity was decreased in the combinations, compared to TFT alone, which is possibly related to an increased inhibition of TS and the induced cell cycle arrest.(15)
Epidermal growth factor receptor inhibitors have been shown to stimulate the induction of apoptosis in various cell types.(15,31) However, in our colon cancer cell line panel, erlotinib itself did not induce cell death, as shown by the sub-G1 analysis. The lack of cell death is supported by the absence of caspase activation following erlotinib treatment. Erlotinib seems to act more on the cell cycle, arresting cells in the G0/G1 phase, which can be mediated by inhibition of pro-survival signaling. As cell death was not involved in the synergistic actions, it is expected that the action of the combinations is more related to an induction of cell cycle arrest, possibly mediated by the induction of DNA damage. Previously it was reported that cell cycle modulation is important for the efficacy of the combination of EGFR-TKIs with cytotoxic agents.(15,37) Cellular damage induced by chemotherapeutic compounds can convert EGFR ligands from growth factors into survival factors for cancer cells that express functional EGFR.(38) In this context, the blockade of EGFR signaling by EGFR-TKI could prevent repair of cellular damage induced by cytotoxic drugs.
Inhibition of EGFR signaling has previously been shown to reduce DNA damage repair.(39) TFT is a known inducer of DNA damage,(5,6) therefore the combined treatment with erlotinib might cause persistent DNA damage induction, for example, TFT-induced DNA damage can not be repaired due to inhibition of the DNA repair mechanisms by erlotinib.(39) This notion is further strengthened by the finding that persistent DNA damage was only observed in EGFR expressing cells. This may also explain why erlotinib alone induced DNA damage, as it also affects basal repair levels. Thus, the synergism found between TFT and erlotinib might be explained by a diminished ability to repair TFT-induced DNA damage lesions.
Thymidine phosphorylase plays a role in apoptosis, cell growth, and angiogenesis.(40) Moreover, TP might be induced after exposure to EGFR-TKIs and cytotoxic agents, which has been related to chemoresistance.(23) TAS-102 consists of TFT combined with thymidine phosphorylase inhibitor.(3) This may be an additional advantage of a clinical combination of TAS-102 with erlotinib. Further investigation is required to elucidate the role of TP in enhancement of responses to EGFR-TKI, especially in vivo. In conclusion, the combination of TFT with EGFR-TKI was synergistic, which was dependent on EGFR expression and probably mediated by cell cycle deregulation and not immediate cell death. Although further research is needed to fully elucidate this promising combination in CRC therapy, the concept of combined molecular targeting of EGFR and TFT seems to present a potential strategy in the field of molecular therapeutics.
This study was financially supported by Tokushima Research Center, Taiho Pharmaceutical Co. Ltd., Tokushima, Japan.
G. J. Peters received a research grant from Taiho Pharmaceutical Co. Ltd; M. Fukushima is an employee of Taiho Pharmaceutical Co. Ltd.