Background
Phase III, open-label RCT. 421 patients with previously untreated PTCL (all subtypes except ALK+ sALCL). Romidepsin is an HDAC inhibitor with single-agent activity in R/R PTCL. Ro-CHOP was designed to test whether adding romidepsin to standard CHOP could improve outcomes in frontline PTCL — a disease where CHOP alone achieves only ~30–40% long-term remissions.
Interventions and follow up
Arm A: Romidepsin 12mg/m² IV days 1 and 8 + CHOP (cyclophosphamide 750mg/m² + doxorubicin 50mg/m² + vincristine 1.4mg/m² [max 2mg] + prednisone 100mg days 1–5) q21d × 8 cycles
Arm B: CHOP q21d × 8 cycles (same doses as above)
Primary endpoint: Progression-free survival
mFollow up: ~6 years (final analysis)
Arm B: CHOP q21d × 8 cycles (same doses as above)
Primary endpoint: Progression-free survival
mFollow up: ~6 years (final analysis)
Results
mPFS: 12.0 vs 10.2 months, HR 0.79, 95% CI 0.62–1.005, P=.054 — missed significance
mOS: Similar between arms (no significant difference)
TFH-phenotype subgroup (exploratory): mPFS 19.5 vs 10.6 months, HR 0.703, P=.039
mOS: Similar between arms (no significant difference)
TFH-phenotype subgroup (exploratory): mPFS 19.5 vs 10.6 months, HR 0.703, P=.039
Adverse events
Hematologic: Grade 3–4 thrombocytopenia 50% (Ro-CHOP) vs 10% (CHOP); grade 3–4 anemia 47% vs 17%; grade ≥3 neutropenia 71% vs 55%
Infectious: Febrile neutropenia numerically higher with Ro-CHOP — substantial hematologic toxicity increase without efficacy benefit
Infectious: Febrile neutropenia numerically higher with Ro-CHOP — substantial hematologic toxicity increase without efficacy benefit
Conclusions
Adding romidepsin to CHOP failed to improve PFS in unselected PTCL (HR 0.79, P=.054), while substantially increasing hematologic toxicity. An exploratory subgroup suggested possible benefit in TFH-phenotype lymphomas (AITL, PTCL with TFH features), which did not reach prespecified statistical significance.
Key Limitations
Missed primary endpoint despite trending toward benefit (HR 0.79, P=.054). High hematologic toxicity (thrombocytopenia 50%, anemia 47%) without efficacy payoff argues against Ro-CHOP in unselected PTCL. The TFH subgroup finding is exploratory and hypothesis-generating only — it would require a dedicated TFH-enriched Phase III to confirm. PTCL is heterogeneous and unselected trials may mask subgroup benefits.
Clinical Context
Ro-CHOP did not change the standard of care. CHOP (or CHOEP in younger patients) remains the backbone for most PTCL subtypes without CD30 expression. A+CHP (ECHELON-2) is the exception for CD30+ disease. The TFH subgroup finding has spurred interest in azacitidine + CHOP in AITL (Ruan et al, Blood 2023 — showing 88.2% CR in TFH-PTCL). Current frontline approach for most PTCL-NOS: CHOP or CHOEP ± ASCT consolidation.