Perioperative/Adjuvant Only Gemcitabine/Nab-Paclitaxel Misses Mark in Resectable PDAC

Disease-free survival was not reached with perioperative nor adjuvant chemotherapy with gemcitabine plus nab-paclitaxel in patients with resectable pancreatic cancer, although neoadjuvant chemotherapy yielded a numerical improvement in overall survival, according to data from the phase 2 NEONAX trial.

Perioperative or adjuvant gemcitabine (Gemzar) plus nab-paclitaxel (Abraxane) did not significantly improve disease-free survival (DFS) in patients with resectable pancreatic ductal adenocarcinoma (PDAC), although a numerical improvement in overall survival (OS) was observed in the neoadjuvant arm, according to findings from the phase 2 NEONAX trial (NCT02047513).

In arm A, patients received perioperative chemotherapy and arm B received adjuvant chemotherapy after up-front surgery. The treatment resulted in an 18-month DFS rate of 33.3% (95% CI, 18.5%-48.1%) when used perioperatively in Arm A and 41.4% (95% CI, 20.7%-62.0%) when used in the adjuvant setting in Arm B, thus missing the pre-defined DFS rate of 55% in both arms. Additionally, the median DFS in the modified intent-to-treat (ITT) population was 14.1 months (95% CI, 10.2-16.8) in Arm A and 17.0 months (95% CI, 10.9-25.1) in Arm B. A total of 66% of patients in arm A and 42% in arm B went on to receive adjuvant chemotherapy and R0/R1 PDAC resection (Arm B) or at least 1 cycle of neoadjuvant chemotherapy (Arm A).

“Neoadjuvant/perioperative treatment is a novel option for patients with resectable pancreatic cancer. However, the optimal treatment regimen has yet to be defined,” the investigators wrote.

The study also assessed outcomes in the ITT population, in which the perioperative cohort experienced better outcomes vs the adjuvant setting. Data for this group showed a median DFS of 11.5 months (95% CI, 8.8-14.5) in Arm A vs 5.9 months (95% CI, 3.6-11.5) in Arm B, as well as corresponding 18-month DFS rates of 30.8% (95% CI, 18.3%-43.2%) and 19.3% (95%CI, 8.0-30.6%), respectively. Moreover, median OS was 25.5 months (95% CI, 19.7-29.7) in Arm A vs 16.7 months (95% CI, 11.6-22.2) in Arm B. The median distant metastasis–free intervals (DMFI) were 14.5 months (95% CI, 11.9-20.6) and 10.9 months (95% CI, 4.1-20.3) in Arm A and Arm B, respectively.

The trial enrolled 127 patients with histologically or cytologically confirmed resectable PDAC, with an ECOG performance status of 0 or 1. The study took place at 22 centers across Germany from 2015 to 2019. Participants were randomly assigned 1:1 into either Arm A (n = 63) or Arm B (n = 64). The analysis assessed a total of 59 eligible participants in each arm. Most patients in each respective arm were male (57.6% vs 62.7%) and nearly all were White (98.3% vs 100.0%). Median age was 65 years (range, 48-82) and 68 years (range, 41-88) in Arms A and B, respectively. In each respective arm, most patients had pancreatic head tumors (69.5% vs 46.0%), T1/T2 tumors (57.6% vs 46.2%), and cN0 tumors (54.2% vs 47.5%).

Patients in Arm A received 2 pre-operative and 4 postop cycles of treatment for 28 days each consisting of nab-paclitaxel at 125 mg/m2 and gemcitabine at 1000 mg/m2 on days 1, 8, and 15. After neoadjuvant treatment, patients underwent a repeated multiphase spiral CT scan with both oral and intravenous contrast enhancement followed by tumor surgery if there was no disease progression. The recommended time between neoadjuvant chemotherapy and surgery was 2 to 3 weeks after last treatment with neoadjuvant therapy. Arm B received surgery up front with subsequent adjuvant chemotherapy for 6 cycles 12 weeks after surgery.

Most adverse effects (AEs) were grade 3 or lower and could be resolved without specific treatment. The toxicity profiles of the treatment in the neoadjuvant and adjuvant settings did not differ greatly following surgery, aside from higher rates of anemia (7.0% vs 3.5%) and infections (10.5% vs 5.3%) in the neoadjuvant arm. Rates of surgical complication within 30 days following resection were also similar, although a higher rate of infections was seen in Arm A (24.4%) vs arm B (8.7%). Mortality was more common in Arm B with 3 grade 5 AEs occurring vs 1 in Arm A. Nearly all deaths occurred within 30 days after resection.

“Taken together, perioperative and preoperative treatment strategies in resectable pancreatic cancer deserve further investigation particularly with respect to the optimal duration and protocols of the preoperative part of the treatment,” investigators concluded.


Seufferlein T, Uhl W, Kornmann M, et al. Perioperative or only adjuvant gemcitabine plus nab-paclitaxel for resectable pancreatic cancer (NEONAX) - a randomized phase II trial of the AIO pancreatic cancer group. Ann Oncol. Published online October 6, 2022. doi:10.1016/j.annonc.2022.09.161