Follicular lymphoma (FL)

Epidemiology & Pathology

• Incidence: ~20 to 30% of NHL in Western countries (~14,000 new cases/yr in the US); 2nd most common NHL after DLBCL and the prototype low-grade lymphoma. Median age ~7th decade (~65); <10% are under age 40; M=F. Etiology unknown (possible environmental link, more common in the "farm belt").
• Natural history: indolent, waxing-and-waning course; can spontaneously resolve; advanced-stage FL is generally incurable with standard systemic therapy, though localized FL can be cured with radiotherapy. Median survival now approaches ~15 to 20 yrs in the modern era (~11 yrs in earliest eras). Advanced stage (III/IV) in ~75 to 85% at diagnosis (bone marrow involvement = stage IV).
• Cell of origin: germinal center B-cell.
• Pathology: t(14;18)(q32;q21) IGH-BCL2 in ~85%. The translocation juxtaposes BCL2 next to the IGH locus → excessively high BCL2 → impaired apoptosis. Phenotype CD20+, CD10+, BCL6+, BCL2+; usually CD5−, CD23−.
• Diagnosis needs sufficient tissue (excisional biopsy) because architecture matters. On low power the node is either nodular (back-to-back follicles) or diffuse (follicles obliterated by malignant cells); grading depends on assessing this and cell size/nuclear shape.

Grading (conventional scheme, retained by ICC 2022; WHO-HAEM5 groups grades 1 to 3A as classic FL with grading optional and renames grade 3B follicular large B-cell lymphoma)

• In young patients with localized disease that lacks BCL2 expression or t(14;18), consider pediatric-type FL (in adults), FL with 1p36 deletion and/or TNFRSF14 mutation, and large B-cell lymphoma with IRF4/MUM1 rearrangement.
• Variants (WHO 2022): pediatric-type FL (younger, localized, no BCL2 rearrangement/BCL2 negative, highly proliferative, excellent prognosis, do not overtreat), in situ follicular neoplasia (ISFN; incidental, low progression), duodenal-type FL (localized to small bowel, low dissemination risk, often watchful waiting), Diffuse-pattern FL (large localized inguinal mass, CD23+, STAT6-altered, usually lacks IGH::BCL2 rearrangement though BCL2 protein may be positive; 1p36/TNFRSF14 variable); good prognosis.

Clinical Presentation

• Painless lymphadenopathy; often asymptomatic at diagnosis.
• Stage III/IV in ~75 to 85%, but can be observed if asymptomatic.
• BM involvement common (~50%); extranodal sites (GI, skin, occasionally).
• Constitutional symptoms uncommon (<20%), unlike DLBCL.

Workup & Prognostic Indices

• Excisional biopsy preferred (architecture matters for grading); PET-CT (highly FDG-avid) for staging; BM biopsy standard. Staging by Lugano criteria (from the Cotswold-revised Ann Arbor system).
• FLIPI-1 (mnemonic "NoLASH"): Nodal areas >4, LDH > ULN, Age >60, Stage III/IV, Hemoglobin <12 g/dL. Score 0 to 5 → low/intermediate/high risk. Despite being derived pre-rituximab, FLIPI retains prognostic significance.
• FLIPI-2: age >60, β2-microglobulin ↑, BM involvement, longest diameter of largest node >6 cm, Hb <12.
• m7-FLIPI: incorporates 7 mutations (EZH2, MEF2B, EP300, ARID1A, FOXO1, CARD11, CREBBP) for high-risk identification.
• Key point: FLIPI is prognostic but does NOT tell you when to treat; the GELF criteria do (see below).
• POD24 (progression within 24 mo of chemoimmunotherapy): ~20% of patients; strong adverse prognostic factor (5-yr OS ~50% vs ~90% for non-POD24; Casulo JCO 2015;33(23):2516). Identifies patients to consider for novel therapy / HSCT.

GELF / NCCN Criteria for Treatment (High vs Low Tumor Burden)

Frontline Treatment: Localized (Stage I/II, ~15 to 25%)

• Involved-site/involved-field RT for stage I to II, grade 1 to 2, potentially curative and gives excellent local control. Lowry trial: 24 Gy = 40 to 45 Gy (no PFS/OS difference); FORT trial (Hoskin Lancet Oncol 2021): 24 Gy superior to 4 Gy for local control, no OS difference. Consider what body region is irradiated (try to avoid neck) and patient age.
• Alternatives: watchful waiting, single-agent rituximab, or chemoimmunotherapy all give similar excellent outcomes (~90% OS at 5 yrs by any approach; Lymphocare registry analysis, Friedberg JCO 2012). R-monotherapy may be the best-balanced option.

Frontline Treatment: Advanced Stage

• No treatment indication (GELF-negative, asymptomatic): watch & wait. Multiple RCTs show no OS benefit from early treatment. RP3 trial (observation vs R-induction vs R-induction + R-maintenance): R + maintenance had best PFS but all arms had the same ~95% OS at 5 yrs. Rituximab monotherapy for asymptomatic patients improves PFS but not OS.
• Low tumor burden with an indication: give rituximab × 4 and observe, re-treating with rituximab × 4 at progression (RESORT trial, Kahl JCO 2014): R-maintenance improved PFS but no difference in time-to-treatment-failure or OS.
• High tumor burden / symptomatic (GELF+):
  • Bendamustine + rituximab (BR): preferred first-line (StiL NHL1, Rummel Lancet 2013): PFS BR > R-CHOP, OS equal, and BR better tolerated.
  • R-CHOP: option, especially for FL3A.
  • R-CVP: acceptable for grade 1, 2, 3A; bendamustine generally preferred.
  • Lenalidomide + rituximab (R2 / R-squared): RELEVANCE (superiority design) compared R2 with rituximab-chemotherapy, mostly R-CHOP (~72%; BR ~23%, R-CVP ~5%); R2 did not demonstrate superiority, with similar CR (55% vs 58%) and PFS (HR 1.10). It was not an equivalence trial (R2 has longer duration, ~18 vs 6 mo, and is more expensive). Toxicity: thromboembolic rates similar between arms; R2 with more rash/cutaneous reactions, R-chemo (mostly R-CHOP) with more neutropenia and febrile neutropenia.
  • Obinutuzumab + chemo (G-CHOP / G-bendamustine): GALLIUM trial (Marcus NEJM 2017; PMID: 28976863) improved PFS over rituximab + chemo (HR 0.66) but no OS benefit; preferred 1L by some, especially in higher-risk disease.
  • Maintenance rituximab × 2 yrs (PRIMA, Salles Lancet 2011): improved PFS, time to next treatment, and time to next chemotherapy, but not OS (10-year OS ~80% in both arms; PRIMA long-term follow-up), so it is not mandatory (and B-cell-depleting maintenance can raise infection risk, a concern highlighted in the COVID-19 era).

Transformation to Aggressive Lymphoma

• Because most FL (~85%) already carry t(14;18)/BCL2, acquiring a MYC rearrangement can produce a double-hit lymphoma. Cumulative risk of histologic transformation ~2 to 3% per year (declining in the modern era); ~30% lifetime.
• Biopsy is mandatory at progression. Suspect transformation with rapid or asymmetric nodal growth, B symptoms, ↑ LDH, ↑ FDG SUV on PET, or new extranodal involvement.
• Risk factors: poor performance status, high LDH, B symptoms, FL grade 3A, high FLIPI, del1q, del6q, +2, +3q, +5, loss of B2M, increased T-reg cells, and TP53/PIM1/B2M mutations. Notably, ~75% of patients with early progression after BR have transformed disease; PRIMA showed worse prognosis for transformation than for FL progression.
• Treatment (guided by prior therapy and transformed histology): no prior chemo → R-CHOP; prior anthracycline: if refractory to first-line or relapse <12 months, second-line axi-cel or liso-cel is preferred; salvage chemo plus auto-HSCT remains appropriate for later chemosensitive relapse; prior bendamustine → R-CHOP ± auto-HSCT. If histology is HGBL/double-hit or triple-hit, treat accordingly (DA-EPOCH-R, etc.).

Relapsed / Refractory FL

• Re-approach as for new FL, deciding by high vs low tumor burden. There are no data on optimal sequencing. General options: chemo/auto-HSCT, allo-HSCT, radioimmunotherapy (ibritumomab tiuxetan, category 2B), lenalidomide + rituximab, and rituximab monotherapy. Two archetypes: POD24 (favor auto-HSCT) and double-refractory FL (resistant to alkylators and anti-CD20; favor CAR-T or allo-HSCT).
• R2 (lenalidomide + rituximab): AUGMENT (Leonard JCO 2019) improved PFS over rituximab alone in R/R indolent NHL.
• Tafasitamab + R2: inMIND, mPFS 22.4 vs 13.9 mo (HR 0.43) vs R2; FDA Jun 2025 for R/R FL.
• Tazemetostat (EZH2 inhibitor): FDA Jun 2020 for R/R FL after ≥2 lines if EZH2-mutated, or any genotype after no satisfactory alternative (PMID: 32563850).
• PI3K inhibitors: idelalisib, copanlisib (IV pan-class I, most active against PI3K-alpha and delta isoforms), duvelisib, umbralisib; all four FL accelerated approvals FDA-withdrawn (duvelisib Dec 2021, idelalisib Feb 2022, umbralisib May 2022, copanlisib Mar 2024); reasons differed by drug (class toxicity with infections and deaths, OS safety signal for umbralisib, failed or unfulfilled confirmatory trials, e.g. CHRONOS-4 for copanlisib).
• CAR-T (R/R FL, ≥2 lines of therapy):
  • Axi-cel (Yescarta): FDA Mar 5, 2021; ZUMA-5 (Jacobson Lancet Oncol 2022): ORR ~94%, CR ~79%; PFS and DOR durable, benefit maintained in POD24.
  • Tisa-cel (Kymriah): FDA May 27, 2022; ELARA (Fowler Nat Med 2022): ORR ~86%, CR ~69%; 12-mo PFS ~67% in R/R FL without transformation.
  • Liso-cel (Breyanzi): FDA May 15, 2024; TRANSCEND FL (Morschhauser ASH 2023): ORR ~96%, CR ~94%.
• Bispecifics (CD20×CD3):
  • Mosunetuzumab (Lunsumio): FDA Dec 22, 2022; first FDA-approved bispecific for FL; IV, 3L+ (GO29781, Budde Lancet Oncol 2022): ORR ~80%, CR ~60%; fixed 8 or 17 cycles.
  • Epcoritamab (Epkinly): FDA Jun 26, 2024; SC bispecific, 3L+ (EPCORE NHL-1, Linton ASH 2023). Epcoritamab + R2: FDA Nov 18, 2025 for R/R FL after ≥1 prior line (EPCORE FL-1: PFS HR 0.21, ORR 89% vs 74% vs R2).
  • Odronextamab (Ordspono): CD20×CD3; CRL issued Mar 2024 (confirmatory trial enrollment) and again Aug 2025 (site inspection); EU approved Aug 2024 for R/R FL/DLBCL after ≥2 lines; not FDA approved in the US.
• Auto-HSCT: consolidation in chemosensitive R/R FL, especially POD24. Allo-HSCT: select young fit, multiply relapsed patients.

High-Yield Pearls

• t(14;18) IGH-BCL2 in ~85%; defining cytogenetics (impaired apoptosis).
• Stage III/IV asymptomatic: watch & wait; no OS benefit from early treatment.
• FLIPI does not decide when to treat; GELF does.
• BR > R-CHOP for grade 1 to 2 FL (StiL NHL1); R2 is a chemo-free alternative with similar CR and PFS (RELEVANCE did not show superiority and was not an equivalence trial).
• POD24: progression within 24 mo; strong adverse prognostic factor and ~75% of early progressors after BR are transformed, so rebiopsy.
• FL grade 3B = DLBCL behavior (CD10−, lacks t(14;18)); treat as aggressive lymphoma.
• Tazemetostat: EZH2 inhibitor for R/R FL.
• Mosunetuzumab (Dec 2022): first FDA-approved bispecific for FL.
• Pediatric-type FL: localized, BCL2 negative, excellent prognosis; do not overtreat.
• Always rebiopsy rapidly enlarging or PET-asymmetric nodes to rule out transformation.