Background: Perioperative chemotherapy (CTX) in resectable pancreatic ductal adenocarcinoma (PDAC) is still not considered standard of care and data are limited. The NEONAX trial examined gemcitabine (Gem) plus nab-paclitaxel (nab-P), in the perioperative or adjuvant therapy of resectable PDAC (NCCN criteria). Methods: NEONAX is a prospective, randomized phase II trial with two independent experimental arms. 127 resectable PDAC patients in 22 German centers were randomized 1:1 to perioperative (2 pre- and 4 postoperative cycles, arm A) or adjuvant (6 cycles, arm B) of Gem (1000mg/m2) and nab-P (125mg/m2) on days 1,8,15 of a 28-day cycle. Results: We previously reported the primary endpoint disease free survival (DFS) at 18 mo. in the modified intention-to-treat (ITT)-population (defined as R0/R1 resected pts. that either started neoadjuvant (A) or adjuvant (B) CTX. The pre-defined DFS rate of 55% at 18 mo. was not reached in both arms (A: 32.2%, B: 41.4%). Here we present the final results of the secondary endpoints median overall survival (mOS), pN0-resection rate, perioperative morbidity/mortality and safety in the ITT-population. Most common grade ≥3 treatment emergent adverse events in the safety population were neutropenia (arm A 21.1%, arm B 12.3%), fatigue (arm A 8.8%, arm B 5.3%) and anemia (arm A 10.5%, arm B 1.8%). The most frequent post-/perioperative complications of all grades in pts. undergoing resection were infections (arm A: 24.4%, arm B: 8.8%), pancreatic fistulas (arm A: 14.6%; arm B: 13.3%) and bleedings (arm A: 9.7%; arm B: 6.7%). Perioperative mortality was 2.4% in the neoadjuvant and 6.7% in the upfront surgery setting. The median number of resected lymph nodes was comparable in both arms (A: n = 21, B: n = 26). The pN0-resection rate was 33.3% in the neoadjuvant/perioperative arm A and 29.5% in the upfront surgery arm B. R0 resection rates were 87.8% in arm A and 67.4% in arm B, respectively. Median OS as a key secondary endpoint in the ITT population was 25.2 mo. in arm A and 16.7 mo. for upfront surgery, a difference of 8.5 mo. This difference corresponds to a mDFS of 11.5 mo. in arm A and 5.9 mo. in arm B. 91.5% of pts. in arm A started and 84.7% completed neoadjuvant CTX but only 42.4% of pts. in arm B started adjuvant CTX. Conclusions: Perioperative treatment with Gem/nab-P was well tolerated and showed an encouraging mOS of 25.2 mo., this is well in the range of the data in SWOG 1505 (23.6 mo.) or PREOPANC (15.7 mo.). The corresponding mOS in the upfront surgery arm was 16.7 mo. The 8.5 mo. difference may be explained by the fact that many pts. in arm B did not receive adjuvant treatment whereas the vast majority of pts. in arm A completed at least preoperative CTX. Neoadjuvant/perioperative treatment is a promising novel option for pts. with resectable PDAC. The optimal treatment regimen is subject of current clinical trials. Clinical trial information: NCT02047513.