Abstract Background: Prior clinical studies in the first and second line setting showed radioimmunotherapy (RAIT) is a promising therapy for pancreatic cancer that avoids the side effects of further chemotherapy. This multicenter study evaluated the contribution of low radiosensitizing doses of gemcitabine (GEM) to fractionated doses of 90Y-clivatuzumab tetraxetan in patients with metastatic pancreatic ductal cancer after having received at least 2 prior systemic therapies. Methods: Fifty-eight patients (33 males, 25 females; median age 63.5 years), 1.6 median years from diagnosis and with a median of 3 (2-7) prior treatments, were randomized to Arm A (N=29, 4-week cycles: 200 mg/m2 GEM, weekly, combined with 6.5 mCi/m2 90Y-clivatuzumab tetraxetan, weekly the last 3 weeks) or Arm B (N=29, 3-week cycles: 6.5 mCi/m2 90Y-clivatuzumab tetraxetan alone, once-weekly), repeating cycles after 4-week delays. Safety and efficacy were evaluated. Results: None of the patients had infusion reactions, and as expected, cytopenias (predominantly thrombocytopenia) were the only significant toxicities, but mostly transient and manageable with infrequent hematologic support and little evidence of increased infection or bleeding. Patients terminated treatment cycles due to disease progression or clinical deterioration, not treatment toxicity. Fifty-three patients (27 Arm A, 26 Arm B, 91% overall) completed ≥1 full treatment cycle and thus were evaluable for efficacy, with 23 (12 Arm A, 11 Arm B; 40%) receiving multiple cycles, including 7 (6 Arm A, 1 Arm B; 12%) given 3-7 cycles. Two patients in Arm A had PRs by RECIST criteria. Karnofsky performance status (90-100 v 70-80), number of prior therapies, and tumor burden estimates (summed length of index lesions, serum CA 19-9 levels) correlated with overall survival (OS), but appear balanced between arms. Kaplan-Meier median OS was 3.9 months (1.0-16.7) in Arm A v 2.8 months (0.9-9.4) in Arm B (hazard ratio 0.54, 95% CI: 0.27-0.87; P=0.020, log-rank). The median OS for Arm A v Arm B increased to 7.9 v 3.4 months with multiple cycles (P= 0.004) and 3 patients in Arm A still being observed (11 – 17 months). Conclusions: This randomized trial demonstrated the feasibility of performing clinical studies in metastatic pancreatic cancer patients after having at least 2 prior therapies (3rd line and beyond). With significant survival advantage and favorable safety profile, fractionated RAIT with 90Y-clivatuzumab tetraxetan and low-dose GEM appears promising in this difficult population, supporting Phase 3 studies of this combination now being initiated. Citation Format: Vincent J. Picozzi, Ramesh K. Ramanathan, Maeve A. Lowery, Allyson J. Ocean, Edith P. Mitchell, Bert H. O'Neil, Michael J. Guarino, Paul R. Conkling, Steven J. Cohen, Nathan Bahary, Richard C. Frank, Tomislav Dragovich, Benjamin B. Bridges, Marie Lee, Ronald L. Korn, Neeta Pandit-Taskar, Stanley J. Goldsmith, Charles M. Intenzo, Arif Sheikh, Timothy C. Manzone, Michael L. Miller, Michael Yu, Judith M. Joyce, Edward B. Strauss, Susan Passalaqua, Ronald V. Dorn, III, Michael J. Anderson, Michael Holt, Fadi S. Braiteh, Fa-Chyi Lee, Thomas E. Gribbin, Donald A. Richards, Alexander N. Starodub, Wegener A. William, Eileen M. O'Reilly, Daniel D. Von Hoff, David M. Goldenberg. Final results of a randomized phase Ib study of fractionated 90Y-clivatuzumab tetraxetan in patients with metastatic pancreatic cancer having at least two prior therapies. [abstract]. In: Proceedings of the AACR Special Conference on Pancreatic Cancer: Innovations in Research and Treatment; May 18-21, 2014; New Orleans, LA. Philadelphia (PA): AACR; Cancer Res 2015;75(13 Suppl):Abstract nr B98.