Conditional progression free survival (PFS) in patients with multiple myeloma (MM) after upfront autologous stem cell transplantation (ASCT): A secondary CIBMTR analysis

Abstract Background: Progression-free survival (PFS) with upfront (1L) multiple myeloma (MM) therapy measures time from diagnosis to progression or death but does not account for the duration a patient has already remained relapse-free. Conditional progression-free survival (cPFS) provides a dynamic, time-updated assessment of progression risk. For newly diagnosed transplant-eligible MM, maintenance therapy is typically continued until progression or unacceptable toxicity. In this context, cPFS may clarify how baseline prognostic factors influence risk over time and help identify patients who might benefit most from maintenance de-escalation. Methods: We screened all publicly available CIBMTR MM datasets (N=12) and identified one study (MM19-01) that reported both induction and maintenance therapy after 1L ASCT. We included all patients who underwent 1L ASCT between 2013 and 2018 with complete PFS data. We estimated cPFS (t∣s) as the probability of remaining progression-free for an additional t months, given survival without progression for at least s months, using the Kaplan-Meier method. We calculated cPFS at 12-, 18-, 24-, 30-, and 36-months after 1L ASCT. We used piecewise Cox non-proportional hazards models with time-varying coefficients to assess the impact of baseline variables on PFS across these discrete intervals. Results: A total of 1,131 patients were included: 20% had ISS stage 3 disease, 38% had high-risk cytogenetics [t(4;14), t(14;16), t(14;20), del17p, +1q, double hit MM] at diagnosis, 81% received VRD induction (19% VCD), 75% received melphalan 200 mg/m2, and 86% received maintenance therapy post-ASCT (77% lenalidomide-based, 9% PI/other, 14% no maintenance). The median PFS (95% CI) after 1L ASCT was 41 (38–49) months. Median cPFS (95% CI) at 12-, 18-, 24-, 30-, and 36-months post-ASCT was 53 (46–61), 60 (53–NR), 66 (59–NR), 73 (65–NR), and 76 (69–NR) months. For all patients, the 12-month cPFS (95% CI) at 12-, 18-, 24-, 30-, and 36-months post-ASCT was 90% (87–92), 82% (78–86), 82% (77–86), 78% (73–83), and 82% (76–87), respectively. Patient subgroups showed distinct cPFS patterns over time. For standard-risk MM, the 12-month cPFS (95% CI) at 12-, 18-, 24-, 30-, and 36-months post-ASCT was 83% (79–87), 85% (80–89), 82% (76–87), 78% (71–84), and 82% (74–88); for high-risk MM, the corresponding estimates were 74% (67–80), 77% (68–83), 80% (70–87), 79% (67–87), and 87% (75–94), with estimates converging around 24 months post-ASCT. For ISS stage 3, the 12-month cPFS (95% CI) at 12-, 18-, 24-, 30-, and 36-months post-ASCT was 72% (63–79), 79% (69–86), 75% (63–84), 69% (54–80), and 78% (60–88); these estimates were consistently lower than for ISS stage 1/2 (83% [78–86], 83% [78–87], 85% [79–89], 81% [73–86], 82% [74–88]). Patients who received VRD vs. VCD induction, or maintenance vs. no maintenance, had overlapping cPFS estimates at all key time points. High-risk cytogenetics were significantly associated with PFS only in the first 0–12 months post-ASCT (HR 2.0 [1.4–2.9]); HRs for 12–18, 18–24, 24–30, 30–36 months: 1.6 (1.0–2.7), 1.7 (0.9–3.1), 1.6 (0.7–3.6), 0.9 (0.4–2.0), and 1.4 (0.8–2.5), respectively. Similarly, day +100 response of CR or better was associated with improved PFS during 0–12 and 12–18 months (HR 0.3 [0.2–0.5] and 0.6 [0.4–0.9]), but not thereafter: HR for subsequent 18–24, 24–30, 30–36 months: 0.7 (0.4–1.3), 0.5 (0.3–1.2), 1.3 (0.6–2.6), and 0.8 (0.5–1.3). Conclusions: cPFS estimates for patients with MM remained stable at 12-, 18-, 24-, 30-, and 36-months following 1L ASCT, supporting the current standard of maintenance therapy until progression or toxicity. Adverse prognostic factors (e.g., not achieving CR) tended to lose significance after the first 12–18 months post-ASCT. Patients with high-risk MM who remained progression-free for the first ~24 months had PFS outcomes comparable to those with standard-risk disease. These findings highlight the importance of dynamic risk assessment and may guide personalized decisions on the duration and intensity of maintenance therapy after ASCT.