Adding nab-paclitaxel to neoadjuvant gemcitabine does not improve the disease-free survival in patients with surgically resected pancreatic cancer
The addition of nab-paclitaxel to gemcitabine previously resulted in a significant survival improvement in patients with metastatic pancreatic cancer. In contrast, a phase III study evaluating the same combination in the adjuvant treatment of patients with surgically resected pancreatic cancer proved to be negative as nab-paclitaxel plus gemcitabine did not improve the disease-free survival compared to gemcitabine alone.
Background
Patients with pancreatic cancer face a very dismal prognosis with a 5-year overall survival (OS) rate of less than 10%. Several clinical trials have established that adjuvant chemotherapy improves the OS of patients with surgically resected pancreatic cancer. In the presented phase III APACT trial, researchers investigated whether the addition of nab-paclitaxel (nab-P) to gemcitabine (Gem) could further improve the outcome of these patients. The choice for this combination of cytotoxic agents was based on data from the phase III MPACT trial that demonstrated a significantly longer OS in metastatic pancreatic cancer patients treated with nab-P plus Gem compared to Gem alone (median OS 8.7 vs 6.6 months; p<0.001). Of note, at the time of the study design, Gem monotherapy was still the standard of care in the adjuvant treatment of patients with surgically resected pancreatic cancer. However, since APACT was launched, the ESPAC-4 trial demonstrated a significantly longer OS with Gem + capecitabine compared to Gem alone in this setting (median OS: 28.0 vs 25.5 months; p=0.032), while the PRODIGE 24 trial did the same for mFOLFIRINOX (median OS: 54.4 vs 35.0 months; p=0.003). As such, Gem + capecitabine and mFOLFIRINOX are now considered as the preferred adjuvant treatment options for patients with pancreatic cancer.
In the phase III, multicenter, open-label, randomized APACT trial, patients with resected pancreatic adenocarcinoma (T1-3, N0-1, M0) were randomly assigned to 6 cycles of nab-P (125 mg/m2 qw 3/4) + Gem (1,000 mg/m2 qw 3/4) or Gem (1,000 mg/m2 qw 3/4) alone. The adjuvant therapy was initiated as early as possible after adequate recovery but no later than 12 weeks after the surgery. The primary endpoint of the study was independently-assessed disease-free survival (DFS), while OS and safety were the main secondary objectives.
Results
In total, 1,226 patients were screened of whom 866 were randomized. The main reason for the screening failures was the presence of persistent or metastatic disease on CT or because of CA19-9 levels exceeding 100 U/mL. Of the 432 patients randomized to nab-P + Gem, 429 (99%) and 287 (66%) started and completed the therapy, respectively. In the Gem arm, 423 of the 434 randomized patients initiated the therapy (97%) and 310 completed the full 6 cycles (71%). The median age of patients in the trial was 64 years and 60% had an ECOG performance status (PS) of 0 (40% ECOG PS 1). An R0 resection was obtained in 76% of patients and the baseline CA19-9 level was 13.65 U/mL. Almost three quarters of patients (72%) were node positive at baseline. The median treatment duration was identical for both treatment arms at 24 weeks, but more patients in the nab-P + Gem arm had to have at least one dose reduction (63% vs 50% with Gem alone). Overall, 59% of the nab-P + Gem patients received nab-P dosing in cycle 6.
The APACT trial did not meet its primary endpoint as the median independently-assessed DFS was similar for nab-P + Gem (19.4 months) and Gem (18.8 months) (HR 0.88 [95% CI 0.729-1.063]; p=0.1824). In a prespecified sensitivity analysis looking at the investigator-assessed DFS, nab-P + Gem did seem to be significantly better than Gem alone (median DFS: 16.6 vs 13.7 months; HR 0.82 [95% CI 0.694-0.965]; p=0.00168), but in this light it needs to be noted that the concordance rate between disease recurrence by independent and investigator review was only 77%. The interim data with respect to OS (data maturity 68%) suggests an OS benefit for nab-P + Gem over Gem alone, but this difference was only borderline significant (median OS: 40.5 vs 36.2 months; HR 0.82 [95% CI 0.680-0.996]; p=0.045). An analysis of the OS in prespecified subgroups indicates that the benefit of adding nab-P to Gem is somewhat more pronounced in patients with moderately differentiated tumors (HR 0.69 [95% CI 0.54-0.89]), in node-positive patients (HR 0.79 [95% CI 0.637-0.971]) and in patients with an R1 resection (HR 0.70 [95% CI 0.518-0.943]).
The small benefit in OS with nab-P + Gem did come at the cost of increased toxicity. In fact, in the nab-P + Gem arm, 86% of patients experienced at least 1 grade 3-5 adverse event as compared to 68% with Gem alone. In the experimental arm 41% of patients suffered at least one serious adverse event versus 23% in the Gem arm. The difference in the incidence of grade 3-5 adverse events between both treatment arms was mainly driven by the higher rate of grade ≥3 neutropenia (58% vs 48%), anemia (15% vs 8%), peripheral neuropathy (15% vs 0%) and fatigue (10% vs 3%).
Conclusions
Adding nab-P to Gem does not prolong the DFS in patients with surgically resected pancreatic cancer. Consistently with other trials the survival with Gem monotherapy was markedly higher than what was seen in historical studies. This is probably the result of better initial patient selection and of the availability of more effective therapies upon disease recurrence. The safety profile of nab-P + Gem was consistent with what was reported in the MPACT trial. The OS subgroup analysis seemed to suggest a more pronounced effect of adding nab-P to Gem in node positive patients and in patients with moderately differentiated tumors or an R1 resection. These observations warrant further study.
Reference
Tempero M, Reni M, Riess H, et al. APACT: phase III, multicenter, international, open-label, randomized trial of adjuvant nab-paclitaxel plus gemcitabine (nab-P/G) vs gemcitabine (G) for surgically resected pancreatic adenocarcinoma. Presented at ASCO 2019; abstract 4000.
