Follicular lymphoma (FL) is the most common indolent B-cell non-Hodgkin lymphoma, arising from germinal center B cells harboring the t(14;18)(q32;q21) translocation. This article covers FL pathobiology, WHO grading, staging, prognostic scoring, and the evolving treatment landscape.
Follicular lymphoma (FL) is the second most common B-cell NHL in the United States and Western Europe, representing approximately 20–25% of all new NHL cases. It occurs predominantly in adults, with a median age of 60 years at diagnosis and a slight female predominance (F:M ratio ~1.3:1). FL is rare in Asia and developing countries, suggesting environmental and genetic influences on its epidemiology. In recent decades, survival has substantially improved due to rituximab-based chemoimmunotherapy, with 10-year overall survival exceeding 70–80% in early studies.
FL arises from germinal center B cells that become "frozen" at the GC stage — unable to exit via normal differentiation into plasma cells or memory B cells. Histologically, FL replicates the follicular architecture of the lymph node, with neoplastic follicles composed of centroblasts and centrocytes in varying proportions (the basis of WHO grading). The neoplastic B cells express GC markers: BCL6, CD10, and HGAL (germinal center B-cell-associated lymphoma protein).
Importantly, t(14;18) is detectable by PCR in the blood of ~70% of healthy adults — indicating that this translocation occurs normally during GC reactions but requires additional cooperating mutations to progress to overt FL. This observation, combined with the long natural history of FL, points to a multi-hit model of lymphomagenesis.
Beyond t(14;18), FL harbors a characteristic set of recurrent somatic mutations identified by whole-exome and whole-genome sequencing studies:
| Gene | Frequency | Function | Effect of Mutation |
|---|---|---|---|
| KMT2D (MLL2) | ~85% | H3K4 histone methyltransferase | Loss-of-function; epigenetic dysregulation; most common FL mutation |
| CREBBP | ~65% | Histone acetyltransferase (HAT) | Loss; impaired p53 acetylation and activation; immune evasion |
| EZH2 | ~25% | H3K27 histone methyltransferase (PRC2) | Y641 gain-of-function; H3K27me3 increase; silences differentiation genes |
| EP300 | ~15% | Histone acetyltransferase | Loss; cooperates with CREBBP loss |
| TNFRSF14 (HVEM) | ~50% | BTLA/CD160 ligand; T-cell suppression reversal | Loss; impairs immunological synapse with T cells; immune evasion |
| BCL2 | ~80% (translocation) | Anti-apoptotic BCL-2 family protein | Overexpression; blocks apoptosis |
| BCL6 | ~15% | GC master TF | Mutation/translocation — disrupts autoregulation |
| CARD11 | ~10% | NF-κB signaling scaffold | Gain-of-function; constitutive NF-κB activation |
| RRAGC | ~17% | mTORC1 activating GTPase | Gain-of-function; nutrient-independent mTORC1 signaling |
The predominance of histone-modifying gene mutations (KMT2D, CREBBP, EZH2, EP300) has classified FL as an "epigenetic disease." These mutations collectively alter the histone modification landscape, disrupting normal GC transcriptional programs governing cell fate. Tazemetostat (EZH2 inhibitor) targets EZH2 Y641-mutant FL specifically.
FL is uniquely dependent on its microenvironment. The immune cell infiltrate within FL follicles includes Tfh-like CD4+ T cells (which provide survival signals analogous to GC Tfh help), Tregs (which suppress anti-tumor immunity), macrophages, and FDC networks. Gene expression studies have defined two microenvironment gene signatures that predict prognosis:
TNFRSF14 (HVEM) loss — common in FL — disrupts the HVEM–BTLA inhibitory axis, impairing T-cell immune surveillance of neoplastic GC B cells.
FL is graded histologically based on the number of centroblasts per high-power field (HPF; defined as 0.159 mm²) by the 2022 WHO Classification of Haematolymphoid Tumours:
| Grade | Centroblasts/HPF | Centrocyte Pattern | Clinical Behavior |
|---|---|---|---|
| Grade 1 | 0–5 | Predominantly centrocytes | Indolent; rarely curable with chemotherapy |
| Grade 2 | 6–15 | Mix centrocytes & centroblasts | Indolent; similar to grade 1 |
| Grade 3A | >15 | Centrocytes still present | Indolent but treated more aggressively |
| Grade 3B | >15 | Solid sheets of centroblasts; no centrocytes | Aggressive; treated like DLBCL; BCL6 translocation common; t(14;18) often absent |
Grades 1 and 2 are clinically indistinguishable and are often managed identically. Grade 3A may have a slightly more aggressive course. Grade 3B FL is treated with R-CHOP (rituximab + cyclophosphamide, doxorubicin, vincristine, prednisone) analogously to DLBCL.
FL staging follows the Lugano Classification (2014), which modified the Ann Arbor staging system. Staging is performed with PET/CT (18-FDG PET), which has replaced CT-only staging due to superior sensitivity for bone marrow involvement and disease extent.
| Stage | Definition | Frequency at Diagnosis |
|---|---|---|
| I | Single lymph node region or single extralymphatic site | ~10% |
| II | Two or more nodal regions on same side of diaphragm | ~20% |
| III | Nodal regions on both sides of diaphragm; ± splenic involvement | ~20% |
| IV | Disseminated extranodal involvement (bone marrow, liver, lung) | ~50% |
The Follicular Lymphoma International Prognostic Index 2 (FLIPI-2) assigns one point each for: age >60 years, hemoglobin <12 g/dL, longest diameter of largest lymph node >6 cm, bone marrow involvement, and serum β2-microglobulin above normal. Risk groups: Low (0 factors) — 3-yr PFS 79%; Intermediate (1–2 factors) — 3-yr PFS 51%; High (≥3 factors) — 3-yr PFS 19%.
The m7-FLIPI integrates 7 gene mutations (EZH2, ARID1A, MEF2B, EP300, FOXO1, CREBBP, CARD11) with FLIPI score and Eastern Cooperative Oncology Group (ECOG) performance status to predict failure-free survival after chemoimmunotherapy.
One of the most feared complications of FL is histologic transformation (HT) to diffuse large B-cell lymphoma (DLBCL) — a rapidly aggressive lymphoma requiring immediate intensive treatment. HT occurs at a rate of approximately 2–3% per year, cumulating to ~30% at 10 years. It is suspected when a patient with known FL develops rapid lymph node growth, B symptoms (fevers, night sweats, weight loss), elevated LDH, or extranodal involvement.
Post-transformation prognosis has historically been poor (median OS ~1–2 years), but patients who are rituximab-naive and achieve complete response with R-CHOP followed by autologous stem cell transplantation (ASCT) have better outcomes. In the CAR-T era, CD19-directed CAR-T therapy (axicabtagene ciloleucel, lisocabtagene maraleucel) has shown efficacy in transformed FL.
Radiation therapy (RT) to involved sites (24–30 Gy) is the treatment of choice for truly limited stage I–II FL and can achieve durable remissions in 40–50% of patients at 10 years. This represents the closest approach to cure in FL. For patients with bulky stage II or comorbidities precluding RT, watchful waiting or rituximab monotherapy may be considered.
Asymptomatic, low-tumor-burden advanced FL is managed with watchful waiting (observation) — landmark BNLI and Stanford trials demonstrated no survival benefit from early treatment versus deferred treatment. Criteria for initiating therapy include: GELF criteria (large tumor bulk ≥7 cm, cytopenia from marrow involvement, B symptoms, pleural effusion, organ compression, rapid progression).
The two main first-line regimens for treatment-requiring advanced FL are:
After successful induction therapy, rituximab maintenance (375 mg/m² every 2 months for 2 years) extends PFS by approximately 2–3 years compared to observation (PRIMA trial: 3-yr PFS 75% vs 58%). Overall survival benefit has not been demonstrated in all trials. Obinutuzumab maintenance after G-CHOP induction also improves PFS.
FL is characterized by a relapsing-remitting course — most patients experience multiple remissions and relapses over their lifetime. Each relapse is generally manageable, but POD24 (progression of disease within 24 months of first-line chemoimmunotherapy) identifies a high-risk subgroup with inferior overall survival (~50% at 5 years vs ~90% in POD24-negative patients).
The ZUMA-5 trial demonstrated that axicabtagene ciloleucel (axi-cel) achieves an ORR of 92% and CR rate of 74% in relapsed/refractory FL after ≥2 prior lines. Median duration of response exceeded 22 months. Axicabtagene ciloleucel is now FDA-approved for FL grade 1–3A after ≥2 prior therapies. Lisocabtagene maraleucel (liso-cel) has also shown efficacy in FL (TRANSCEND FL).
CD20×CD3 bispecific antibodies recruit T cells to kill CD20-expressing FL cells. Mosunetuzumab (step-up dosing SC) is FDA-approved for relapsed/refractory FL after ≥2 prior lines — ORR 80%, CR 60%, with notable durability. Epcoritamab and glofitamab are also in trials for FL. These agents cause cytokine release syndrome (CRS) and neurotoxicity (ICANS) at rates generally lower than CAR-T therapy.