Efficacy of second-line chemotherapy after a first-line triplet in patients with metastatic colorectal cancer

Background Exposing patients with metastatic colorectal cancer (mcrc) to all three active chemotherapeutic agents (oxaliplatin, irinotecan, fluorouracil) has improved survival. The benefit of second-line chemotherapy after a first-line triplet is not clearly defined. We evaluated the efficacy of second-line chemotherapy in patients who had received first-line triplet therapy. Methods The medical records of patients treated on a prospective trial of first-line triplet therapy were reviewed for second-line treatment. Univariate and multivariate analyses were performed to establish factors of prognostic significance. Results Of the 53 patients who received first-line triplet therapy, 28 (53%) received second-line chemotherapy [13 men; 8 with a colon primary; mutant KRAS in 10, wild-type in 15, and unknown status in 3; Eastern Cooperative Oncology Group performance status (ps) of 1 in 16 patients, ps 2 in 3, ps 3 in 2, and unknown in 7; involved organs: liver in 17 patients, lung in 16, and peritoneum in 8]. Second-line chemotherapy consisted of xelox or folfox in 13 patients, xeliri or folfiri in 12, and single-agent irinotecan in 3. Concurrent bevacizumab was given in 16 patients (57%), and cetuximab, in 2 (7%). Median survival was 28.0 months [95% confidence interval (ci): 22.8 months to 33.2 months] for patients receiving second-line therapy and 23.0 months (95% ci: 13.2 months to 32.8 months) for those not receiving it. Best response was partial in 6 patients (21%), stable disease in 11 (39%), and progressive disease in 11 (39%). Median progression-free survival was 4.8 months (95% ci: 2.4 months to 9.6 months), and overall survival was 15 months (95% ci: 9.6 months to 20.4 months). Conclusions Second-line chemotherapy after first-line triplet therapy in mcrc is feasible and suggests efficacy comparable to that reported for second-line therapy after a doublet, regardless of the agent used.


INTRODUCTION
The treatment of metastatic colorectal cancer (mcrc) had changed considerably since the early 2000s.The most important milestones were the introduction of irinotecan 1 , oxaliplatin 2 , bevacizumab 3 , and monoclonal antibodies against the epidermal growth factor receptors 4,5 .Doublet chemotherapy regimens using either oxaliplatin or irinotecan-folfox (leucovorin, fluorouracil, oxaliplatin), folfiri (leucovorin, fluorouracil, irinotecan)-are considered standard-of-care options for the first-line treatment of mcrc 6 .
Compared with doublet chemotherapy regimens, triplet regimens have been associated with improved progression-free survival (pfs) and overall survival (os) by some investigators [7][8][9][10] .Second-line therapy after failure of a first-line single-agent or doublet regimen has shown efficacy, with improved survival [11][12][13] .The benefit of second-line chemotherapy after failure of a triplet regimen has been less extensively investigated 14,15 .We previously published the results of our phase i/ ii trial of a triplet consisting of capecitabine, oxaliplatin, and irinotecan with bevacizumab (xeloxiria) in patients with advanced crc 16 .In that study, 53 patients received xeloxiria, with 4% achieving a complete response, and 60% achieving a partial response.Median pfs was 16 months, and median os was 28 months.Toxicity was high, with grades 3 and 4 toxicity rates of 36% for diarrhea, 21% for vomiting, and 17% for fatigue.Here, we report the efficacy of second-line chemotherapy in patients for whom the first-line triplet regimen failed.

METHODS
Medical records and case report forms for patients with mcrc treated with first-line xeloxiria in a prospective clinical trial (NCT01311050 at http://ClinicalTrials.gov/) were reviewed for second-line chemotherapy.All patients received first-line xeloxiria, which consisted of oral capecitabine at a dose of 1000 mg/m 2 twice daily for 14 days, intravenous oxaliplatin 130 mg/m 2 over 2 hours on day 1, intravenous irinotecan 150 mg/m 2 over 90 minutes on day 1, and intravenous bevacizumab 7.5 mg per kilogram body weight over 30 minutes on day 1.Cycles were repeated every 21 days.After the first 30 patients, the dose of capecitabine was reduced to 800 mg/m 2 twice daily for 14 days because of excessive toxicity.After 5-8 cycles, patients achieving a response or stable disease were maintained on capecitabine and bevacizumab until progression, undue toxicity, or death.
Data collected for patients who received second-line chemotherapy included performance status at the start of second-line chemotherapy, complete blood count, serum carcinoembryonic antigen (cea) and alkaline phosphatase (alp), number and location of metastatic sites, response to first-line chemotherapy, pfs on first-line chemotherapy, KRAS status (NRAS was not available), chemotherapy-free interval (cfi) before second-line therapy, and second-line regimen (number of cycles, response, pfs, and os).

Survival Analysis and Statistics
For all patients receiving second-line therapy, imaging was reviewed for assessment of response according to recist (Response Evaluation Criteria in Solid Tumours), version 1.1 17 .Measurement of pfs began at the start date of second-line chemotherapy and ended at the date of first documented disease progression or death from any cause.Measurement of os began at the start date of second-line chemotherapy and ended at the date of death from any cause.If a patient was not known to have died, survival was censored at the date of last contact.Tabulation and statistics were performed in the SAS statistical software application (version 9.4: SAS Institute, Cary, NC, U.S.A.).The Kaplan-Meier method was used to calculate pfs and os.Calculation of p values used the log-rank test, and results were considered statistically significant if equal to or less than 0.05.
Univariate and multivariate Cox regression analyses were performed for pfs and os, including factors that might have an impact (age, sex, performance status, serum cea, serum alp, KRAS status, tumour site, liver involvement, number of organs involved, response to first-line therapy, pfs for first-line therapy, cfi before second-line therapy, and type of second-line chemotherapy).

Ethics Considerations
This retrospective study was approved by the hospital's institutional review board with waiver of consent, given that most of the patients had died by the time of the study.Patient confidentiality was maintained throughout the study.

Patient Characteristics
The first-line triplet trial of xeloxiria enrolled 53 patients with mcrc.Results have been reported 16 .Of those patients, 28 (53%) received second-line therapy.The most common reasons for not proceeding to second-line chemotherapy were either no progression or poor performance status.A small proportion did not proceed because of toxicity from first-line chemotherapy or withdrawal of consent.Table i summarizes the characteristics of the patients who received second-line chemotherapy.

Efficacy of Second-Line Chemotherapy
Of the 28 patients receiving second-line chemotherapy, 6 (21%) achieved a partial response, and 11 (39%) achieved stable disease, for a disease control rate of 61%.The remaining patients experienced disease progression.No patient achieved a complete response.At a median follow-up of 28 months, median pfs was 4.8 months [95% confidence interval (ci): 2.4 months to 9.6 months] and os was 15 months (95% ci: 9.6 months to 20.4 months; Figure 1).Median os was 28 months (95% ci: 22.8 months to 33.2 months) for patients who received second-line chemotherapy and 23 months (95% ci: 13.2 months to 32.8 months) for those who did not receive second-line chemotherapy (log-rank p = 0.69, Figure 2).Irinotecan-based chemotherapy was given to 15 patients, and oxaliplatin-based chemotherapy, to 13 patients.The pfs and os for irinotecan-based second-line therapy were 3.6 months (95% ci: 1.2 months to 8.4 months) and 18 months (95% ci: 1.2 months to 21.6 months); for oxaliplatin-based chemotherapy, they were 8.4 months (95% ci: 2.4 months to 12 months) and 14.4 months (95% ci: 12 months to not reached).

Prognostic Factors
Table ii shows the pfs and os durations for patient subgroups by potential prognostic factors.Despite a trend toward better results for patients who were female, who had elevated serum alp, who had wild-type KRAS, who achieved a partial response on first-line chemotherapy, or who received oxaliplatin-based rather than irinotecanbased chemotherapy, no difference in pfs or os reached statistical significance.Only normal serum cea (compared with elevated cea) and cfi greater than 3 months (compared with 3 months or less) at the time of second-line chemotherapy were associated with significantly longer pfs and os in univariate analysis.Cox regression analysis showed that elevated alp was an independent prognostic factor for improved pfs.On the other hand, female sex was the only independent prognostic factor for improved os;

DISCUSSION
The benefit of exposing patients with crc to all three active chemotherapeutic agents (oxaliplatin, irinotecan, fluorouracil) has been shown to improve survival 18 .Accordingly, the value of second-line chemotherapy with any of those agents in patients who receive all three drugs in the first line has been less clear.Because many of our patients did not receive second-line chemotherapy (for a number of reasons), it is difficult to prove directly in this study that second-line chemotherapy improves survival in patients who previously received triplets.However, at least indirectly, it appears that second-line chemotherapy is as effective in this patient population as it is in patients who receive second-line chemotherapy after initial doublet therapy 6 .
In the present study, the pfs and os for patients unselected for RAS status receiving second-line chemotherapy   after triplet therapy were 4.8 months (95% ci: 2.4 months to 9.6 months) and 15 months (95% ci: 9.6 months to 20.4 months) respectively.Those results are very similar to results in many contemporary trials of second-line therapy after doublet chemotherapy.In the velour trial, the pfs was 4.6 months in the standard arm and 6.9 months in the aflibercept arm.Similarly, os was 12.1 months in the standard arm and 13.5 months in the aflibercept arm 19 .In the tml trial (bevacizumab after progression), pfs was 4.1 months in the standard arm and 5.7 months in the bevacizumab arm; os was 9.8 months and 11.2 months respectively 20 .Many other trials have also reported results of secondline chemotherapy 11,12,21,22 .The outcome of second-line chemotherapy after a first-line triplet has also been reported by gono (Italy's Gruppo Oncologico Nord Ovest) 14,15 .In their report, second-line chemotherapy after folfoxiri in 136 patients resulted in an overall response rate of 23%.Median pfs and os were 5.9 months and 13.2 months respectively-a result similar to that reported here.In an exploratory subgroup analysis, re-treatment with folfoxiri compared with a second-line doublet was associated with longer pfs (8.2 months vs. 6.3 months, p = 0.003; hazard ratio: 0.61) and os (19.3 months vs. 14.0 months, p = 0.02; hazard ratio: 0.57).Compared with folfoxiri or a doublet, single-agent chemotherapy or fluoropyrimidine plus mitomycin C was associated with a significantly lower response  The univariate subgroup analysis suggested lower efficacy for second-line chemotherapy in patients with a cfi of less than 3 months.That finding is similar to results reported by the gono group and likely represents patients who are truly chemotherapy-refractory, which is probably why cfi did not remain significant in the multivariate analy sis.In the gono report, the os for patients with a cfi of 3 months or less was 5.9 months, compared with 13.4 months for patients with a cfi of 3-6 months 15 .Interestingly, elevated alp and female sex emerged as factors of prognostic significance for pfs and os respectively.Although women have fared slightly better than men in most second-line trials [23][24][25][26] , the difference has not reached statistical significance, and it was not considered by the gono group.We believe that a pooled analysis of individual patient data from similar trials might provide an answer.
Limitations of our study include the small number of patients and the retrospective nature of the analysis.On the other hand, our study is the only one outside the Italian experience to address the issue of second-line chemotherapy after triplet chemotherapy.

CONCLUSIONS
Our data support the feasibility of second-line chemotherapy in patients for whom a first-line triplet regimen has failed.Efficacy with the second-line chemotherapy in that situation is similar to the efficacy seen after a first-line doublet.

FIGURE 1
FIGURE 1 Kaplan-Meier plots of (A) overall survival (OS) and (B) progression-free survival (PFS) in patients treated with second-line chemotherapy after first-line triplet chemotherapy.CI= confidence interval.

FIGURE 2
FIGURE 2 Kaplan-Meier plot of overall survival in patients treated with and without second-line chemotherapy.

TABLE I
Characteristics of 28 patients receiving second-line chemotherapy

TABLE II Univariate
SECOND-LINE CHEMOTHERAPY AFTER FIRST-LINE TRIPLET IN CRC, Bazarbashi et al.

TABLE IV
Multivariate analysis of factors with possible prognostic significance for overall survival

TABLE III
Multivariate analysis of factors with possible prognostic significance for progression-free survival