Long term responders to palliative chemotherapy for advanced biliary tract cancer

Background: Patients with advanced biliary tract cancer (BTC) are often treated with palliative chemotherapy (PC). Standard PC since 2010 is a cisplatin/gemcitabine doublet, with median overall survival (OS) of 11.7 months from the ABC-02 trial. Prior to this, our institutional standard was gemcitabine and fluoropyrimidine. The ABC-02 study used 8 cycles of PC as standard with treatment stopped even in the absence of disease progression, but some patients may benefit from continuing PC longer than 8 cycles. Methods: Patients treated with at least 2 cycles of PC for advanced BTC in Princess Margaret Cancer Centre between 1987 and 2015 were included, and divided into 2 groups for analysis-long-term responders (LTR) who received 9 or more cycles, and controls (2-8 cycles). Data was collected on demographics, clinicopathological features, PC regimen, toxicities, and survival. The primary outcome measure was OS, with secondary analyses including progression-free survival (PFS) and toxicity rates between groups. Results: A total of 382 patients were identified, 123 who met the criteria for LTR and 259 who were included as controls. The baseline demographic and clinical characteristics were similar, although more patients in the control group had gallbladder cancer or extrahepatic cholangiocarcinoma than LTR (P=0.024), and more patients in the LTR group were treated with combination chemotherapy regimens (93% vs. 82% in controls, P= 0.003). The LTR patients had significantly longer PFS (median 13.3 vs. 4.1 months, P <0.001) and longer OS than controls (median 22.1 vs. 9.2 months, P<0.001). In LTR patients, 15% had a break from chemotherapy of 3 months or more and restarted the same regimen. The LTR patients reported higher rates of nausea, cutaneous and hematologic toxicity, but also more frequently went on to receive second-line chemotherapy (47% vs. 33%, P=0.007). In multivariable analysis of OS, LTR, good performance status and intrahepatic site of cancer were associated with better survival. Conclusions: From this institutional dataset, a significant proportion of patients continued chemotherapy past 8 cycles, and appeared to derive benefit from longer duration of treatment. Toxicity rates were higher in this group, but manageable as evidenced by second-line treatment rates. Discontinuation of chemotherapy for reasons other than toxicity or progression may result in loss of disease control and impact survival in this population; these data suggest the use of continued chemotherapy to disease progression in patients with advanced BTC is a favorable option.