Capecitabine and oxaliplatin combined with bevacizumab are feasible for treating selected Japanese patients at least 75 years of age with metastatic colorectal cancer

Colorectal cancer ranks worldwide as the third and fourth most common cancer in women and men, respectively, and the median survival of patients with metastatic colorectal cancer (mCRC) treated with best supportive care is approximately 6 months [1–3]. Treatment outcomes are improved considerably by newly developed chemotherapeutic agents and regimens. For example, treatment using 5-fluorouracil (5-FU) plus irinotecan, oxaliplatin, or both combined with targeted agents extends median overall survival (OS) to approximately 30 months [4, 5]. Current guidelines recommend that first-line treatment for patients with mCRC should include doublet chemotherapy plus a targeted agent, if tolerated [6].

The prodrug capecitabine is activated by a unique mechanism that exploits the high activity of thymidine phosphorylase in malignant tissue that generates 5-FU preferentially in tumor tissue [7]. Capecitabine undergoes a three-step enzymatic conversion, and the final stage is catalyzed by thymidine phosphorylase, which is significantly more active in tumor tissue compared with healthy tissue [7, 8]. Oral delivery of capecitabine simplifies chemotherapy and provides convenient outpatient therapy, because it avoids the complications and discomfort associated with intravenous administration and permits prompt discontinuation of treatment when toxicity occurs [9].

Combining capecitabine with oxaliplatin (XELOX) is advantageous for the reasons as follows: synergistic effects, no overlapping toxicities, easy to administer, and outpatient management [10–13]. Randomized phase III trials demonstrate that outcomes using first-line XELOX are comparable with those achieved using continuous infusion of 5-FU and folinic acid combined with oxaliplatin (FOLFOX) [14, 15]. Moreover, combined with bevacizumab, a recombinant humanized version of a mouse monoclonal antibody against human vascular endothelial growth factor, XELOX achieves significantly improved progression-free survival (PFS) compared with chemotherapy alone [16–18].

The average age of the population is steadily increasing in many developed countries, particularly because of improvements in public health, nutrition, disease prevention, early detection, and continued progress in medical research [19]. The increase in patients’ ages presents the medical community with new challenges. For example, more than 30 % of patients with newly diagnosed CRC are aged at least 75 years [20]. Since the progressive reduction of functional reserve that occurs in various organs with ageing might increase the susceptibility of the elderly to adverse effects, clinical trials for elderly patients with mCRC have been conducted and tolerability of UFT/leucovorin, XELOX, capecitabine plus bevacizumab and S-1 plus bevacizumab were evaluated [21–26]. Yet, the safety and efficacy of XELOX plus bevacizumab for elderly patients remains to be determined because earlier large clinical trials limited eligibility to individuals <70 or 75 years owing to frail health [16, 17, 27]. Recently, a phase II trial (BECOX study) found that XELOX combined with bevacizumab is effective and well tolerated by patients aged ≥70 years with mCRC who reside in Spain [28]. However, insufficient evidence is available to establish the safety and benefit of XELOX plus bevacizumab for Japanese patients with mCRC included in this age group. Moreover, lack of robust evidence of the new treatment described above may subject patients of advanced age to more conservative and less effective treatments. For example, older patients are more likely to receive monotherapy instead of combination therapy that does not include agents that target specific molecules [29–31].

Therefore, the aim of the present study was to evaluate the feasibility of XELOX plus bevacizumab for selected Japanese patients with mCRC aged ≥75 years.

Robust evidence from the TREE 1 (XELOX) and TREE 2 (XELOX plus bevacizumab) (TREE1/2) randomized clinical trials shows that XELOX combined with bevacizumab offers survival benefits to patients with mCRC [33]. Unfortunately, insufficient evidence is available to insure the safety and benefits of combined treatment with XELOX and bevacizumab for patients aged ≥75 years, that were often excluded from randomized trials, allegedly because of frail health or because they represented a minority of enrolled patients [16–18]. Feliu et al. conducted a recent phase II trial (BECOX study) in Spain and demonstrated that XELOX plus bevacizumab was effective and well tolerated by patients with mCRC aged ≥70 years [28]. Here we designed a multicenter open-label phase II trial to evaluate the safety and efficacy of XELOX plus bevacizumab for Japanese patients aged ≥75 years with mCRC. The doses of capecitabine, oxaliplatin, and bevacizumab were determined with reference to the TREE1/2 trials [33], although the median age of patients enrolled in these studies was 62 years.

The overall frequency of grade 3/4 AEs, including hematologic and nonhematologic events, is generally consistent with those of the TREE 1/2 trials, an earlier phase I/II trial conducted in Japan and the BECOX study conducted in Spain [28, 33, 34]. The most characteristic finding here was that the incidence of grade-1 neuropathy reached 83.3 %, and that of grades-3/4 neuropathy was 13.9 %. Cumulative neuropathy represents one of the major problems related to long-term therapy using oxaliplatin-containing regimens for patients with mCRC, which is the main driver for trying to limit the dose of oxaliplatin [12, 14, 28]. The frequency (13.9 %) of G3/4 neuropathy encountered here was higher compared with those reported by earlier studies of Western cohorts (Table 4), although dose reduction and discontinuation of oxaliplatin was strictly defined in the study protocol [13, 28, 35]. A pilot study evaluating the safety of XELOX plus bevacizumab conducted in Japan reported a 17 % frequency of neuropathy G3/4 17 %, indicating that the frequency of severe neuropathy induced by XELOX plus bevacizumab differs between Western and Japanese patients [34]. Haller et al. showed the regional differences in tolerability of XELOX between the United States, East Asia, and the rest of the world [36]. Japanese patients experienced fewer G3/4 AEs during XELOX treatment compared with those from other regions, but no detailed data for neuropathy was provided.

Further, there remains room for discussion about the survival benefit of adding oxaliplatin. For example, in the AVEX study that evaluated capecitabine plus bevacizumab versus capecitabine alone in patients with mCRC aged ≥70 years, the OS of those treated with capecitabine plus bevacizumab is similar OS to that our present study (20.7 months) [22]. Further, the FOCUS2 trial that compared capecitabine plus oxaliplatin with capecitabine alone, found no significant benefit of adding oxaliplatin [23]. Considering the high prevalence of neuropathy here, the benefit of adding oxaliplatin to capecitabine combined with bevacizumab for older Japanese patients with mCRC should be evaluated in clinical trials involving a large number of patients.

During the present study, one patient died because of treatment-related intracerebral bleeding. Although most AEs associated with bevacizumab (hypertension, proteinuria, and bleeding) are manageable, they infrequently lead to death. The patient had normal blood-clotting function as defined by the eligibility criteria, and the onset of intracerebral bleeding occurred after seven cycles of the protocol dose of XELOX plus bevacizumab. However, the overall safety profile of XELOX combined with bevacizumab for patients aged ≥75 years was similar to those of previous clinical trials [27, 33]. From our experience, we propose to monitor neurological signs on each visit and perform cerebral imaging on low threshold in symptomatic patients. In the present study, the incidence of AEs was independent of patients’ sex, age, and BMI. In contrast, low baseline CCr (<64 ml/min) was associated with the frequency of severe AEs, suggesting that baseline CCr should be considered as a determinant of the suitability of treating older patients with XELOX plus bevacizumab. However, further studies of a larger cohort are required.

Our trial achieved response and disease control rates of 55.6 and 91.7 %, respectively. The primary endpoint, median PFS, was 11.7 months (95 % CI 8.0–13.4 months), and the median OS was 22.9 months (95 % CI 17.6–33.0 months). The median PFS in the TREE 2 trial, the earlier Japanese phase I/II trial, and the BECOX study were 10.3, 11.0 and 11.1 months, respectively [21, 28, 33]. The median OS of patients was 22.9 months in our present study, which is somewhat shorter compared with large studies of younger populations. For example, an OS of approximately 29 months was reported by the FIRE-3 and CALGB/SWOG 80405 trials [4, 5]. In contrast, an earlier study of XELOX combined with bevacizumab for Western patients with mCRC aged 75 years demonstrated that OS was 20.4 months [28].

Folprecht et al. analyzed the differences in efficacy of 5-FU-based chemotherapy between age groups >70 years and <70 years with mCRC, and concluded that elderly patients benefit at least to the same extent from palliative chemotherapy with 5-FU compared with younger patients [37]. Recently, Lieu et al. analyzed the large database of the ARCAD Clinical Trials Program and evaluated primary age effects and interactions with sex and PS [38]. They demonstrated that greater age was associated with poorer OS and PFS among treated patients with mCRC independent of sex and PS [38]. The main reason for the survival differences between our study and those of the FIRE-3 and CALGB/SWOG 80405 trials might be accounted for by the age of the patients rather than regional differences, and our results can be considered to reveal a reasonable outcome for patients aged ≥75 years [4, 5].

The present study included some limitations as follows. The relatively small sample size precluded subgroup analysis of age, second-line treatment, and renal function. We selected patients according to strict eligibility criteria to ensure consistency with those of younger individuals. Therefore, these criteria may not be applicable to routine clinical practice. In addition, serial data were unavailable for blood cholesterol, triglyceride, and glucose concentrations that are influenced by capecitabine. The discussion might be limited due to lack of data on RAS/BRAF status. No elderly specific evaluation was conducted though the comprehensive geriatric assessment would have been of high value to learn about factors that are specific to the older patient population which could affect treatment outcome. Because the study protocol had no provisions regarding the second-line treatment, the detailed information of second-line treatment is unavailable. We were unable to determine the survival benefit of XELOX plus bevacizumab because this was a single-arm study.

Our results indicate that XELOX combined with bevacizumab was well tolerated by selected Japanese patients aged ≥75 years with mCRC. Therefore, XELOX plus bevacizumab should not be withheld from these patients because of age alone. The survival benefit of this regimen must be determined by further controlled clinical trials.