Background
For decades the prognosis of advanced pancreatic cancer has been dismal with 5 year survival rates of approximately 2 and 4 % for metastatic and locally advanced unresectable pancreatic cancer, respectively [
1]. Gemcitabine had been the standard of care for metastatic pancreatic cancer since 1997 based on modest survival benefit compared to bolus 5-fluorouracil (5.6 vs. 4.4 months, P = 0.002) and clinical benefit response (23.8 vs. 4.8 %, P = 0.002) [
2]. In 2010 a new standard of care emerged when the combination regimen FOLFIRINOX (oxaliplatin, irinotecan, fluorouracil, and leucovorin) was shown to significantly improve the survival of fit patients with metastatic pancreatic cancer compared to gemcitabine (11.1 vs 6.8 months, HR = 0.57, P < 0.001) [
3]. More recently the two drug combination of gemcitabine with nab-paclitaxel (G + Nab-P) was also shown to increase survival compared to gemcitabine alone in previously untreated patients with metastatic pancreatic cancer (8.5 vs 6.7 months, HR = 0.72, P < 0.001) [
4].
Despite the increasing use of FOLFIRINOX for advanced pancreatic cancer since 2010, there are no published studies of efficacy or tolerability of second line regimens in patients who had received first line FOLFIRINOX. The use of gemcitabine with G + Nab-P is an attractive option in FOLFIRINOX-treated patients, given its activity in previously untreated patients with metastatic pancreatic cancer [
3]. However, the optimum dose, tolerability, and efficacy of second-line G + Nab-P after first line FOLFIRINOX are unknown. In this single institution retrospective review, we evaluated the tolerability and efficacy of second line G + Nab-P in patients with metastatic or locally advanced unresectable pancreatic cancer who had received first line FOLFIRINOX.
Methods
We conducted a retrospective review of all patients with locally advanced unresectable or metastatic pancreatic cancer who were treated at Yale’s Smilow Cancer Center between December 2011 and December 2013 with second line G + Nab-P after previous therapy with FOLFIRINOX. This study was approved by Yale IRB. The data for each patient was obtained by review of the patients’ electronic medical records and cross-sectional imaging studies. For each patient, we assessed their demographics, performance status, prior therapy, doses of G and Nab-P, the relative dose intensities of G and Nab-P (the proportion of the administered cumulative dose relative to the planned cumulative dose of G 1000 mg/m2 and Nab-P 125 mg/m2 days 1, 8, and 15, every 4 weeks), hematologic toxicities, best response by RECIST (version 1.1), time to treatment failure (TTF), and overall survival (OS). All response assessment scans were reviewed by the investigators with independent on-site radiologists. TTF was defined as time from first treatment with G + Nab-P to disease progression, unacceptable treatment toxicity (per treating physician’s assessment), patient preference, or death. OS was defined as time from first treatment with G + Nab-P to death. The median TTF and OS were determined using Kaplan–Meier method. We also determined the overall response rate and the median relative dose intensities of G and Nab-P.
Discussion
FOLFIRINOX and G + Nab-P have been shown to be more effective than gemcitabine monotherapy as first-line therapy in patients with previously untreated metastatic pancreatic cancer, extending survival to 11.1 and 8.5 months, respectively, compared to approximately 6.7 months with gemcitabine [
2‐
4]. Despite the increasing and widespread use of FOLFIRINOX in fit patients with advanced pancreatic cancer, there is a paucity of data regarding second line regimens after first line FOLFIRINOX. The use of the FOLFIRINOX and G + Nab-P regimens in sequence is an attractive option to maximize disease control and survival.
In this single institution retrospective review, second line G + Nab-P demonstrated modest activity in patients with advanced pancreatic cancer who had previously been treated with FOLFIRINOX. The response rate of 17.9 % exceeds the response rate of single agent gemcitabine in the first line setting [
2], and 46 % of patients achieved disease control (PR + SD) for at least 7 weeks. The TTF and OS of 2.8 and 5.2 months, respectively, are similar to the benefit seen with gemcitabine in the first line setting.
These results represent the largest reported experience with G + Nab-P after FOLFIRINOX in the literature. There are two published case reports of the use of G + Nab-P after FOLFIRINOX [
5,
6]. In addition, two smaller retrospective series with 12 and 7 patients treated with second line G + Nab-P after prior FOLFIRINOX have been reported [
7,
8].
Given the toxicities of FOLFIRINOX, concerns have been raised that patients will not tolerate a second line regimen, particularly, a doublet regimen. The majority of the patients in this study had received FOLFIRINOX for at least 6 months and were still able to tolerate this second line doublet. Notably, all patients in this study started at a reduced dose of Nab-P (≤100 mg per square meter), and thirteen patients (46.4 %) started G at a reduced dose as well. Ten and fifteen patients required subsequent dose reductions of Nab-P and G, respectively, primarily for hematologic toxicities, and 39.3 % of patients required growth factor support. The median dose intensities of G and Nab-P, 62.4 and 57.5 %, were substantially lower than reported for G and Nab-P in previously untreated patients (75 % for G and 81 % for Nab-P) [
4]. Despite substantial myelosuppression, we observed no episodes of neutropenic fever or life-threatening anemia or thrombocytopenia. Although the grade of neuropathy was not documented in all patients at initiation of G + Nab-P, only one patient discontinued Nab-P due to worsening neuropathy, and this patient had neuropathy at initiation of G + Nab-P. Two patients required dose attenuations because of fatigue.
There has been no established standard of care for second line therapy in advanced pancreatic cancer. Multiple prospective trials examined a number of single agents (5-FU, irinotecan, taxanes, erlotinib, platinum agents) and doublet combinations after disease progression on gemcitabine [
9‐
16]. A meta-analysis of 34 studies published between 2000 and 2012 confirmed modest benefit of second line chemotherapy compared to best supportive care (OS 6 vs. 2.8 months, P = 0.01) [
17]. A randomized phase II study of modified FOLFIRI and modified FOLFOX after gemcitabine showed favorable efficacy and toxicity profiles with median PFS and OS of 8.3 and 16.6 weeks, respectively, in the FOLFIRI arm, and 6.0 and 14.9 weeks, respectively, in the FOLFOX arm [
18]. Thus, the efficacy of G + Nab-P after FOLFIRINOX in our study is similar to what has been reported for second line therapy after less effective first line gemcitabine. This observation suggests that initial treatment with the multi-drug FOLFIRINOX regimen does not limit the use or compromise the efficacy of second line chemotherapy compared to initial treatment with gemcitabine.
Although the efficacy of second line G + Nab-P in our study is similar to what has been reported for second line therapy after gemcitabine, it is substantially lower than the efficacy of G + Nab-P in previously untreated patients (TTF 2.8 v 5.5 months; OS 5.3 v 8.5 months) [
4]. The reductions in dose intensity of both G and Nab-P, greater tumor burden, and increased likelihood of chemotherapy resistance in the second line setting, all may have adversely impacted efficacy compared to G + Nab-P in untreated patients.
The limitations of this study include its retrospective design, the small number of patients identified for analysis, the heterogeneity of the patients with inclusion of locally advanced and metastatic disease, and the lack of consistent dosing or dose modifications. The choice of a second line regimen was at the discretion of the treating physicians; the majority of FOLFIRINOX-treated patients received non-Nab-P containing second line gemcitabine-based regimens during the time period of this analysis, prior to the approval of Nab-P by the Food and Drug Administration for pancreatic cancer. However, all patients who were treated at this academic center between 12/2011 and 12/2013 with second line G + Nab-P after first line FOLFIRINOX were included in this analysis, without any selection bias.
Conclusions
At present, we do not have any validated predictive biomarkers to guide the selection of initial chemotherapy (FOLFIRINOX or G + Nab-P) for patients with advanced pancreatic cancer. Thus, it will be important to explore the optimum sequencing of these regimens, as well as the appropriate dosing, tolerability, and efficacy in prospective trials. The efficacy of G + Nab-P after first line FOLFIRINOX in this retrospective review is encouraging. Our findings demonstrate the feasibility of a sequential strategy of FOLFIRINOX followed by G + Nab-P and support further evaluation of this sequence in a prospective trial to establish the optimum dosing, tolerability, and efficacy.