Lymphoma Trial Design Patterns
Definition
Lymphoma encompasses Hodgkin lymphoma (HL) and non-Hodgkin lymphoma (NHL). NHL is further divided by histology: diffuse large B-cell lymphoma (DLBCL, most common aggressive NHL), follicular lymphoma (FL, most common indolent NHL), mantle cell lymphoma (MCL), marginal zone lymphoma (MZL), T-cell lymphomas (peripheral T-cell, ALCL, AITL), and primary CNS lymphoma. Response is assessed by Lugano 2014 criteria using PET-CT (CT-based response for some indolent histologies).
CTG Phase 3 lymphoma dataset (300 trials): PFS primary 75/300 (25%), EFS primary 42/300 (14%), OS primary ~35/300 (12%), ORR/CR primary ~50/300 (17%).
Key response criteria: Lugano 2014 — CR requires PET-negative (Deauville score 1–3) or CT-based CR (no enlarged lymph nodes). PR requires ≥50% reduction in sum of products of perpendicular diameters (SPD). PD requires ≥50% increase or new lesions. EFS-12/EFS-24 landmark rates are validated surrogates for OS in DLBCL.
Regulatory Position
Aggressive NHL (DLBCL, MCL): EFS (event-free survival) is increasingly used as primary. EFS-24 (event-free at 24 months) is a validated surrogate for OS in DLBCL (Patterson et al. 2019). CR rate (by IRC) supports accelerated approval.
Indolent NHL (FL, MZL): PFS is standard primary for regular approval. Median PFS of 5–10 years in FL means OS is impractical. Large PFS HRs (0.35–0.55) support approval.
Hodgkin Lymphoma: CR rate and PFS primary in relapsed/refractory settings. Brentuximab vedotin (BV) approved based on ORR in relapsed HL.
CAR-T cell therapy (DLBCL): EFS primary (ZUMA-7, TRANSFORM — ide-cel and liso-cel vs. SOC). EFS HR 0.40 (ZUMA-7) supported regular approval.
Status: FDA Cancer Endpoints 2018 = Final; Lugano 2014 criteria
When to Use Each Endpoint
| Histology | Setting | Primary Endpoint | Key Secondary | Notes |
|---|---|---|---|---|
| DLBCL | 1L (R-CHOP vs. novel) | EFS or OS | CR rate by IRC, PFS | POLARIX (pola-R-CHP, PFS primary, HR 0.73) |
| DLBCL | 2L (CAR-T vs. SOC) | EFS | OS, CR rate | ZUMA-7 (axi-cel, EFS HR 0.40); TRANSFORM (liso-cel, EFS HR 0.35) |
| DLBCL | 2L (salvage chemo) | OS or EFS | CR rate, PFS | CORAL (R-ICE vs R-DHAP) |
| Follicular lymphoma | 1L (anti-CD20 + chemo) | PFS | OS, CR rate, time to next treatment | GALLIUM (obinutuzumab, PFS HR 0.66) |
| Follicular lymphoma | Maintenance | PFS | OS | PRIMA (rituximab maintenance, PFS HR 0.55) |
| Mantle cell lymphoma | 1L (BTK inhibitor) | PFS | OS, CR rate | TRIANGLE (ibrutinib + R-CHOP) |
| HL | 1L advanced | PFS or EFS | OS | ECHELON-1 (BV+AVD vs ABVD, PFS HR 0.77) |
| HL | Relapsed/refractory | ORR (CR rate) | PFS, OS | Accelerated approvals for BV, pembrolizumab |
| T-cell lymphoma | 1L | OS or EFS | CR rate | ECHELON-2 (BV+CHP, OS HR 0.66) |
Design Considerations
EFS Definition (Lymphoma-Specific)
EFS in lymphoma includes a broader event list than solid tumor EFS:
Standard lymphoma EFS events:
- Progression or relapse of disease
- Initiation of new anti-lymphoma therapy (not pre-specified protocol therapy)
- Death from any cause
The inclusion of "initiation of new anti-lymphoma therapy" as an event is lymphoma-specific — in solid tumors, new therapy is usually handled as an IE (censoring event for PFS) or treatment policy. In lymphoma, new therapy initiation is clinically meaningful (therapy is changed because the current regimen is failing or has failed) and is incorporated as a composite event.
Lugano 2014 PET-CT Assessment
Response assessment requires:
- Baseline PET-CT before first treatment (within 4 weeks of randomization)
- End-of-treatment (EOT) PET-CT after completion of prescribed treatment (within 6–8 weeks of last dose)
- Deauville score 1–5 for each lesion (1=no uptake, 5=markedly increased uptake)
- CR = Deauville 1–3 (with or without residual mass on CT)
- PD = Deauville 4–5 with new or increased FDG uptake
IRC requirements: Open-label trials must have central PET-CT review (BICR) blinded to treatment. Deauville scoring must be performed by qualified nuclear medicine/radiology reviewers.
EFS-12 / EFS-24 as Surrogate
For aggressive lymphoma (DLBCL):
- EFS-12: Binary outcome — is patient event-free at 12 months? Validated surrogate for OS in RCHOP-treated DLBCL
- EFS-24: Event-free at 24 months — stronger surrogate for long-term OS; patients achieving EFS-24 have OS approaching age-matched general population
Caveat: Surrogacy validated for RCHOP-era data (pre-2020). Novel regimens (CAR-T, BTK inhibitors) may have different EFS-OS relationship — surrogacy should be re-validated.
CAR-T Cell Therapy as Subsequent IE
CAR-T (axi-cel, liso-cel, tisa-cel) is an IE in post-CAR-T maintenance or consolidation trials. In 2L DLBCL trials where CAR-T is the experimental arm:
- Control arm receipt of CAR-T post-SOC failure: Treatment policy for OS; RPSFT as sensitivity if formal crossover designed
- EFS: Composite strategy — progression, initiation of new non-protocol therapy, or death
Intercurrent Events
New Anti-Lymphoma Therapy (Composite Strategy)
In lymphoma EFS, new therapy initiation is an EFS event (not a censoring event). This is the key difference from solid tumor PFS:
- SAP language: "Event-free survival events include: disease progression or relapse per Lugano 2014 criteria, initiation of new systemic anti-lymphoma therapy not pre-specified in the protocol, or death from any cause, whichever occurs first."
Transformation (Indolent to Aggressive Lymphoma)
Histologic transformation of FL to DLBCL is a clinically significant event in FL trials:
- Strategy: Composite — transformation counts as an EFS/PFS event
- Must be centrally confirmed by pathology
- Pre-specify whether transformation without prior RECIST/Lugano progression is an event
HSCT in Lymphoma
Similar to AML: treatment policy (include post-HSCT outcomes in OS and EFS). CAR-T and ASCT are part of the treatment continuum in relapsed lymphoma — censoring distorts the comparison.
Regulatory Precedent
| Trial | Drug | Setting | Primary Endpoint | HR / Result | Approval |
|---|---|---|---|---|---|
| GALLIUM | Obinutuzumab + chemo | 1L FL | PFS by IRC | HR 0.66 | Regular (2017) |
| POLARIX | Pola-R-CHP vs R-CHOP | 1L DLBCL | PFS by IRC | HR 0.73 | Regular (2023) |
| ZUMA-7 | Axi-cel vs SOC | 2L DLBCL | EFS | HR 0.40 | Regular (2022) |
| TRANSFORM | Liso-cel vs SOC | 2L DLBCL | EFS | HR 0.35 | Regular (2022) |
| ECHELON-1 | BV+AVD vs ABVD | 1L HL | PFS | HR 0.77 | Regular (2018) |
| ECHELON-2 | BV+CHP vs CHOP | 1L CD30+ T-cell | OS | HR 0.66 | Regular (2018) |
| PRIMA | Rituximab maintenance | FL maintenance | PFS | HR 0.55 | Regular (2011) |
Limitations and Pitfalls
EFS-OS surrogacy in novel therapy era: EFS-12/24 validated for RCHOP. CAR-T, BTK inhibitors, and bispecific antibodies may have different EFS kinetics (late relapses, distinct resistance patterns) that alter the EFS-OS relationship.
PET-CT inter-reader variability: Deauville scoring has moderate inter-reader agreement. Central review is essential; local Deauville scoring has ~20% discordance with central review.
New anti-lymphoma therapy as composite event vs. IE: The lymphoma-specific practice of counting new therapy initiation as an EFS event differs from solid tumor practice (where it's a censoring IE in hypothetical strategy). This creates cross-indication comparability challenges.
Long indolent lymphoma follow-up: FL trials require 10+ year follow-up for OS. PFS primary at 3–5 years with OS as immature secondary is the practical reality.
Backlinks
- Oncology Endpoint Overview
- DFS and EFS Endpoints
- Response-Based Endpoints (ORR, CR, DOR)
- Intercurrent Events in Oncology Trials
Source: FDA Cancer Endpoints 2018 (Final); Lugano 2014 Classification Compiled from FDA guidance + CTG Phase 3 lymphoma dataset (300 trials) + known Phase 3 trial results