TAS-102 for Treatment of Advanced Colorectal Cancers That Are No Longer Responding to Other Therapies

Worldwide, colorectal cancer is the third most frequently diagnosed cancer in men and the second in women, with over 1.4 million new cases diagnosed in 2012. Approximately 25% of patients have metastatic disease at the time of presentation, and almost 50% will develop metastases eventually (1, 2). A minority of patients with metastatic disease can be treated by surgery, which is the preferred treatment if feasible. In most cases of metastatic colorectal cancer (mCRC), however, resection is not feasible and chemotherapy is the mainstay of palliative treatment for most patients.

Until the beginning of this century, 5-fluorouracil (5-FU) was the only anticancer drug available for systemic treatment of mCRC. Fortunately, the number of available therapies has expanded, and median survival of patients with mCRC has improved markedly since oxaliplatin, irinotecan, bevacizumab (a VEGF inhibitor), cetuximab, and panitumumab (EGFR inhibitors) have been introduced. For mCRC patients who have progressed on these therapies, the National Comprehensive Cancer Network (NCCN) guideline currently lists three options: best supportive care, participation in a clinical trial, or treatment with regorafenib. Regorafenib, a multikinase inhibitor, is the only systemic treatment option that has demonstrated statistically significant survival benefit in a randomized, placebo-controlled trial. The overall survival (OS) benefit of regorafenib was 1.4 months, and 50% of patients developed progressive disease within two treatment cycles, often accompanied by severe toxicity (3). Altogether, this emphasizes the need for additional treatment options for patients who have exhausted all initial therapies.

TAS-102 (Lonsurf; Taiho Oncology) is a novel oral formulation of the combination of trifluorothymidine (trifluridine or TFT) and tipiracil hydrochloride (TPI), a thymidine phosphorylase inhibitor (Fig. 1). TFT was originally synthesized in 1964 (4). Despite promising antitumor activity in breast and colon cancer (5), clinical development was discontinued due to unacceptable toxicity and a poor pharmacokinetic profile. The rapid degradation of TFT by thymidine phosphorylase (TP) led to a mean plasma half-life of TFT of less than 15 minutes (6). Only after a potent TP inhibitor (TPI) became available, clinically meaningful plasma TFT concentrations could be reached (7, 8). Optimal TFT concentrations were obtained with TFT and TPI in a 1:0.5 molar ratio (9). Interestingly, the formulation (named TAS-102) showed antitumor activity against both 5-FU–sensitive and 5-FU–resistant models in preclinical studies (10, 11). Its activity in 5-FU–resistant tumors was confirmed in clinical studies, in which TAS-102 showed antitumor activity in mCRC patients who had exhausted all of the above-mentioned therapies. TAS-102 might hereby represent an important new therapeutic option for patients with treatment-refractory mCRC.

Multiple phase I trials of TAS-102 monotherapy were conducted to determine the maximum tolerated dose (MTD) and optimal dosing schedule (Table 1). The first phase I study enrolled 14 patients with therapy-refractory mCRC. On the basis of a preclinical monkey model, treatment was administered once daily for 14 days in a 21-day schedule. Of note, 50 mg/m²/day was declared the MTD, with neutropenia being the dose-limiting toxicity (DLT). No objective tumor responses were observed, but stable disease (SD) was observed in 29% of patients (12).

Even though no DLTs occurred at a dosage of 50 mg/m²/day, myelotoxicity appeared to be more pronounced in later treatment cycles. This observation seemed to be explained by progressive drug accumulation as observed at the end of every treatment cycle. Two additional phase I studies were therefore initiated to define a dosing interval that would minimize hematologic toxicity. TAS-102 was administered once daily for 5 days per week in both studies, either on a 2-weeks-on, 2-weeks-off schedule (n = 24), or on a 1-week-on, 2-weeks-off schedule (n = 39). The recommended phase II dose for these schedules was 100 and 160 mg/m²/day, respectively. Because the 2-week rest between doses did not decrease blood concentration of FTD, and because of the higher administered dose, the 2-weeks-on, 2-weeks-off schedule was recommended. Consistent with the data from the study by Hong and colleagues (12), neutropenia was the DLT for both dosing schedules. No objective responses were seen, but SD was observed in almost 30% of patients (13).

Because of the limited antitumor activity of TAS-102 in these studies, and because earlier studies suggested improved antitumor activity of FTD when it was administered in multiple daily doses (5, 14, 15), another phase I study was initiated in which TAS-102 treatment was divided over three daily doses (16). In this study, 60 mg/m²/day was declared the recommended phase II dose. Although no objective responses according to RECIST criteria were seen, 5 patients did show radiographic reductions in tumor burden and 60% of patients had SD.

Based on these results, an additional phase I study was initiated in Japan to establish the MTD and optimal phase II dose in Japanese patients (17). Patients were treated with two daily doses for 5 days per week on the 2-weeks-on, 2-weeks-off schedule. Although the MTD was not reached, the dosage of 35 mg/m² twice daily already caused grade 3 neutropenia in 3 of 6 patients and was therefore declared the recommended dose for subsequent studies. No objective tumor responses were observed, but 52% of patients showed SD with a mean duration of 2.5 months.

This study was followed by a double-blind phase II trial of 169 Japanese patients with unresectable chemorefractory mCRC (18). Patients were randomized between placebo or TAS-102, 35 mg/m² twice daily for 5 days per week on a 2-weeks-on, 2-weeks-off schedule, based on the study by Doi and colleagues (17). One patient in the TAS-102 group (1%) achieved a partial response (PR), whereas no objective responses were observed in the placebo group. Median overall survival (OS) was 9.0 months [95% confidence interval (CI) 7.3–11.3] in the TAS-102–treated patients (n = 112) and 6.6 months (95% CI, 4.9–8.0) in the placebo group (n = 57; HR for death, 0.56, P = 0.001; ref. 18).

One more phase I study was initiated to confirm the safety of the 35 mg/m² twice-daily dosage in Western mCRC patients (19). Based on the treatment of 27 patients with chemorefractory mCRC, 35 mg/m² twice daily was indeed declared the recommended dose for Western mCRC patients. SD was reached in 65% of patients with a median PFS and OS of 4.1 and 8.9 months, similar results to those from the Japanese phase II trial.

A double-blind, randomized, placebo-controlled phase III study (RECOURSE trial) was then conducted to assess the efficacy and safety of TAS-102 in a global population of patients with mCRC who were refractory or intolerant to standard-of-care therapy (20). Patients were stratified according to the KRAS status of their tumor, time between first diagnosis of metastasis and randomization (<18 vs. ≥18 months), and ethnicity. OS was the primary endpoint; secondary endpoints included PFS, response rate, disease control rate, and safety. A total of 800 patients were randomized (2:1) to receive TAS-102 plus best supportive care or placebo plus best supportive care. Both treatment groups were comparable with respect to prior systemic therapies and were heavily pretreated, with ≥4 prior treatment lines in 61% of patients. At the time of primary analysis, the median OS was 7.1 months (95% CI, 6.5–7.8) in the TAS-102 group versus 5.3 months (95% CI, 4.6–6.0) in the placebo group, leading to an HR for death of 0.68 (95% CI, 0.58–0.81, P < 0.001). The median PFS was 2.0 months (95% CI, 1.9–2.1) in the TAS-102 group and 1.7 months (95% CI, 1.7–1.8) in the placebo group, with an HR for progression of 0.48 (95% CI, 0.41–0.57, P < 0.001). The survival benefit of TAS-102 was observed across all prespecified subgroups, was irrespective of prior regorafenib use (which became available during the trial), and was validated by multivariate Cox regression analysis. Although the objective response rate did not statistically significantly differ between the TAS-102 and placebo group (1.6% vs. 0.4%, P = 0.29), the disease control rate was higher in the TAS-102 group (44% vs. 16%, P < 0.001). Despite the absence of extensive quality-of-life data, TAS-102 treatment significantly delayed the worsening of ECOG performance status [5.7 months vs. 4.0 months in the placebo group, HR 0.66 (95% CI, 0.56–0.78, P < 0.001)].

To summarize, the OS benefit of TAS-102 in Western and Japanese patients with mCRC refractory or intolerant to standard therapies was demonstrated in the multinational, randomized, double-blind and placebo-controlled phase III RECOURSE study (20). Based on the findings from this study and supported by the almost identical results of the randomized, double-blind and placebo-controlled phase II study in Japanese patients (18), TAS-102 was approved for treatment of mCRC refractory to fluoropyrimidine-, oxaliplatin-, and irinotecan-based chemotherapy, and VEGF and EGFR inhibitors. It should be noted, however, that objective tumor responses have rarely been observed. TAS-102 should therefore be regarded more as a tumor-growth stabilizing drug than as a drug inducing objective tumor shrinkage.

TAS-102 is generally well tolerated. Neutropenia was the major DLT in all phase I studies, for Japanese and Western patients. Most grade 3 and 4 toxicities across all studies were of hematologic origin. Other common toxicities included grade 1–2 gastrointestinal complaints and fatigue (12, 13, 16–19).

In the pivotal phase III study, the overall incidence of adverse events (AE) in the TAS-102–treated group was similar to that of the placebo-treated group (98% vs. 93%). However, grade ≥3 AEs occurred more frequently in the TAS-102 group than in the placebo group (69% vs. 52%). TAS-102–treated patients were more likely to develop grade ≥3 neutropenia (38% vs. 0%), leukopenia (21% vs. 0%), anemia (18% vs. 3%), and thrombocytopenia (5% vs. <1%), as well as grade ≥3 diarrhea (3% vs. <1%), nausea (2% vs. 1%) and vomiting (2% vs. <1%). One TAS-102 treatment-related death due to sepsis was reported. The overall incidence of serious AEs, however, was higher in the placebo group (34% vs. 30%) and incidence of grade ≥3 hepatic or renal dysfunction, anorexia, stomatitis, hand–foot syndrome, and cardiac events (typical toxicities of fluoropyrimidine treatment) was similar between the two groups. Altogether, TAS-102 was very well tolerated. This is further illustrated by the fact that only 4% of the TAS-102–treated patients discontinued treatment because of toxicity, and only 14% required dose reductions. However, due to neutropenia, the start of a new treatment cycle was delayed by ≥4 days in 53% of patients (20).

The key pharmacokinetic parameters of TFT and TPI are shown in Table 2. Following TAS-102 administration, TFT and TPI are rapidly absorbed with mean T_max values of 1 to 2 hours for TFT and 2 to 3.5 hours for TPI (17). No absolute bioavailability study has been conducted with TAS-102 to evaluate the oral bioavailability of TFT and TPI. TFT is rapidly degraded by intestinal and liver thymidine phosphorylase. TPI is an inhibitor of thymidine phosphorylase. The TFT AUC was approximately 100-fold higher following administration of TAS-102 than following administration of TFT alone. The TFT C_max was 70-fold higher for TAS-102 compared with TFT alone (7).

In vitro studies showed that TFT and TPI are not metabolized by CYP enzymes. It was demonstrated that TFT is primarily metabolized by thymidine phosphorylase; 5-(trifluoromethyl) uracil (FTY) is the major inactive metabolite (12). TPI is a specific inhibitor of TPase and consequently an inhibitor of the metabolism of FTD in the intestinal tract and liver. Because of the low absorption of TPI, its activity might be focused in the intestinal tract.

Clinical comparative studies in which patients are randomized between TAS-102 and standard therapies have not yet been performed. Cetuximab and panitumumab have demonstrated single-agent activity in chemorefractory mCRC (21, 22), but only regorafenib has been tested in a similar setting to that of TAS-102, in a global, multicenter, randomized, double-blind, and placebo-controlled phase III study (the CORRECT trial; ref. 3).

As in the RECOURSE trial, all patients in the CORRECT trial were heavily pretreated and refractory to their last line of standard treatment. Criteria for refractory disease In the RECOURSE trial were even more stringent, however, requiring that patients were refractory to their last administration of all standard therapies. In addition, 18% of patients in the RECOURSE trial had previously received and progressed on regorafenib, and survival benefit was maintained regardless of prior regorafenib use. The median PFS and OS in the CORRECT study were 1.9 and 6.4 months in the regorafenib group, versus 1.7 and 5.0 months in the placebo group. These findings are comparable with the PFS (2.0 months for TAS-102 vs. 1.7 months for placebo) and OS benefit of TAS-102 (7.1 months for TAS-102 vs. 5.3 months for placebo) as observed in the RECOURSE study. Both studies reported similar disease control rates (41% for regorafenib vs. 44% for TAS-102).

However, regorafenib has a very different safety profile, with skin toxicity, fatigue, diarrhea, anorexia, oral mucositis, hypertension, and elevated transaminases being the most common toxicities. From the CORRECT and RECOURSE trials, regorafenib seemed to be less well tolerated than TAS-102, with AEs occurring in almost 100% of patients (vs. 93% for TAS-102), dose modifications in 67% (vs. 14%), treatment discontinuation due to toxicity in 10% (vs. 4%) of patients, and 8 deaths that could not be attributed to disease progression in 500 regorafenib-treated patients (vs. one in 533 TAS-102-treated patients). Altogether, the efficacy of TAS-102 for patients with mCRC who have exhausted all treatment options seems to be comparable with that of regorafenib, the only approved drug in this setting. The safety profile is different and seems to be in favor of TAS-102. However, as no comparative studies of regorafenib and TAS-102 have been performed, no direct comparison can be made.

In summary, TAS-102 has clinically meaningful antitumor activity and a manageable safety profile, as demonstrated in a global population of patients with heavily pretreated and treatment-refractory mCRC. Based on the results of the RECOURSE trial (20) and supported by the results of the phase II study by Yoshino and colleagues (18), TAS-102 has been approved in Japan by the FDA and recently by the European Medicines Agency for treatment of mCRC refractory to fluoropyrimidine-, oxaliplatin-, and irinotecan-based chemotherapy, and VEGF and EGFR inhibitors.

Some questions remain to be elucidated. Several studies are currently being conducted to further explore the mechanism of action of TAS-102, and to gain insight into the effects of TAS-102 treatment in patients with renal insufficiency and hepatic impairment (Table 3). Another ongoing topic of investigation is combination treatment. Combined treatment approaches could enhance the antitumor activity of TAS-102 or might help to prevent recurrent disease due to drug resistance. Optimal treatment combinations as well as optimal treatment sequences remain to be established, but results from preclinical and even a few clinical studies are promising (23–29). In addition to exploring combination treatment [e.g., with panitimumab (Clinicaltrials.gov NCT02613221) or bevacizumab (Clinicaltrials.gov NCT02654639)], future studies should try to identify predictive biomarkers indicating benefit or resistance to TAS-102. [18F]fluorothymidine PET imaging, for example, has been suggested as a means to assess the pharmacodynamics of TAS-102 in individual patients (30). The role of TAS-102 in earlier lines of colorectal cancer treatment, as well as its role in treatment of other tumor types, needs to be evaluated to decide whether the benefits derived from TAS-102 can be expanded. One such study looking at TAS-102 plus best supportive care versus placebo plus best supportive care in patients with metastatic gastric cancer is ongoing (Clinicaltrials.gov NCT02500043). While we await these answers, TAS-102 already offers a new opportunity for treatment-refractory mCRC patients. Although TAS-102 is well tolerated, its tumoristatic rather than tumoricidal effect and small OS benefit should be considered before prescribing the drug.

No potential conflicts of interest were disclosed.

Conception and design: D.L. van der Velden, F.L. Opdam, E.E. Voest

Development of methodology: D.L. van der Velden, F.L. Opdam

Acquisition of data (provided animals, acquired and managed patients, provided facilities, etc.): D.L. van der Velden

Analysis and interpretation of data (e.g., statistical analysis, biostatistics, computational analysis): D.L. van der Velden, F.L. Opdam, E.E. Voest

Writing, review, and/or revision of the manuscript: D.L. van der Velden, F.L. Opdam, E.E. Voest

Study supervision: F.L. Opdam