Consensus recommendations from the 2024 International Follicular Lymphoma Scientific Workshop

Abstract

Follicular lymphoma (FL) is the most common indolent non-Hodgkin lymphoma. Although patients with FL have high response rates to therapy, most develop increasingly resistant disease. In addition, transformation into an aggressive lymphoma is associated with unfavorable outcomes. Many novel agents are under investigation, and early clinical data are encouraging. Aligning treatment with the underlying tumor biology and sequencing of therapies remain key clinical challenges. At the Lymphoma Research Foundation’s biannual 2024 Follicular Lymphoma Scientific Workshop, experts convened to discuss the role of chemotherapy in the context of new therapies, the impact of early progression on treatment sequencing, novel end points in clinical trials, disease biology and the tumor microenvironment, and new treatments on the horizon. This report focuses on updates in FL biology, first-line treatment, the role of progression of disease in 24 months, clinical trial design, and redefining cure in FL.

Introduction

Follicular lymphoma (FL) is an indolent B-cell non-Hodgkin lymphoma, in which most patients harbor the t(14;18) chromosomal translocation,¹ resulting in overexpression of the B-cell lymphoma 2 (Bcl-2) protein.² Mutations in chromatin-modifying genes are common features of the disease, and the lymphoma microenvironment (LME) plays a key role in disease pathogenesis.3, 4, 5, 6, 7 Progress in drug development has enabled most patients to achieve an overall survival (OS) of ≥15 years.⁸ However, the natural history of FL can include histologic transformation to a higher-grade subtype, including diffuse large B-cell lymphoma, which is associated with decreased OS.9, 10, 11 Although most patients with FL respond to current standard-of-care treatments, many relapse and require intermittent treatment throughout their lifetime. Consequently, there are unmet patient needs that should be addressed, including the impact of FL and its treatment on quality of life(QOL). Understanding the marked heterogeneity of FL remains a challenge at both biologic and clinical levels. Crossdisciplinary collaboration is critical to better understand the disease and to devise solutions that address existing gaps in patient care.

The Lymphoma Research Foundation, the largest nonprofit organization in the United States exclusively dedicated to lymphoma research and patient advocacy, hosted the biannual 2024 International Follicular Lymphoma Scientific Workshop, funded through the Lymphoma Research Foundation Jaime Peykoff Follicular Lymphoma Initiative. The workshop examined recent updates in FL biology, including advances in the genetics and epigenetics of FL and their interplay with the immune microenvironment. Speakers debated the continued role of chemoimmunotherapy (CIT) as first-line therapy in context of emerging therapies, such as bispecific antibodies (BsAbs). The prognostic and therapeutic value of progression of disease within 24 months of initial therapy (POD24) was addressed. Next, the discussion focused on circulating tumor DNA (ctDNA) and minimal residual disease (MRD) as biomarkers in FL, and their potential roles in clinical trials. New and emerging therapies were also explored. Finally, the panel deliberated on the feasibility of curing FL within the current therapeutic framework. This report reviews these topics, and Table 1 summarizes key recommendations.

Table 1.

Lymphoma Research Foundation workshop recommendations

Recommendations
Despite marked heterogeneity in FL, biologic characteristics are not currently factored into treatment decisions, likely leading to both overtreatment and undertreatment. Identification of predictive biomarkers to guide treatment decisions should be a key goal for ongoing and future first-line clinical trials.
Biologic insights are needed to determine which patients with FL are likely to transform to aggressive lymphoma; these patients should be prioritized for clinical trials using novel approaches.
QOL end points should be considered mandatory for randomized FL trials. Only by integrating both efficacy outcomes and QOL measurements can we gain a fuller understanding of the impact of treatment.
Because current first-line FL treatments achieve durable remissions for many patients, the development of surrogate end points (such as detection of ctDNA) is needed to expedite trial readouts and therapeutic development in FL. Modern ctDNA approaches should be feasible to perform in all, or nearly all, patients with FL.
Extended follow-up in FL clinical trials is critical to understanding long-term outcomes and toxicities, especially for novel treatments, including CAR T cells and BsAbs.
Treatment goals, including the formidable goal of achieving a cure in FL, differ between patients. Aggressive early interventions should be targeted to selected patients.
Treatment goals, including the formidable goal of achieving a cure in FL, differ between patients. Aggressive early interventions should be targeted to selected patients.
Achieving curative strategies for FL will require a comprehensive molecular understanding of the mechanisms driving its heterogeneity, early progression, and transformation.
Increased efforts are needed to ensure that novel therapeutic approaches are accessible to all patients with FL.

FL biology

Over the past decades, significant progress has been made in understanding the biology and pathogenesis of FL, which results from a complex interplay of genetic, epigenetic, immunologic, and microenvironmental factors. However, the remarkable heterogeneity of the disease, both spatial and temporal, and technological limitations, including scarcity of accurate preclinical models and difficulty of obtaining adequate biologic material, pose challenges to achieving a comprehensive view of FL that could inform patient management strategies.

Genetic insights

Genomic analysis of sequential patient samples has allowed us to infer the evolutionary history of FL, revealing that dominant clones at diagnosis, relapse, and transformation originate through divergent evolution from a common mutated precursor cell (CPC).^3,12 Recurrent mutations in epigenetic modifier genes, such as those encoding for acetyltransferases (eg, CREBBP) and methyltransferases (eg, KMT2D and EZH2), have emerged as genetic hallmarks of the disease, which occur early in the CPC, preceding its final clonal expansion into FL or transformed FL (tFL).^3,12 Notably, cells carrying CREBBP mutations can be detected in blood of apparently healthy individuals who later develop FL, up to 12 years before diagnosis.¹³ Therefore, intercepting the CPC is critical to eradicating the reservoir of cells responsible for relapse and transformation.¹⁴ Targeting vulnerabilities specific to the CPC may pave the basis for rationally selected, effective treatments capable of preventing its progression to a more aggressive state. Another unresolved challenge is mapping the full spectrum of functionally relevant genetic alterations in the largely unexplored noncoding genome of tFL, because these regions may harbor pathogenic events that contribute to deregulated expression of driver genes, and may reveal new therapeutic targets. Increasing evidence indicates the preferential enrichment of aberrant somatic hypermutation activity in tFL, targeting noncoding sequences with potential regulatory functions.^12,15,16 Notably, a recent whole-genome sequencing study developed a classifier based on coding and noncoding mutations that divided FL into 2 genetically distinct subgroups showing a remarkable difference in time to transformation, with the so called diffuse large B-cell lymphoma–like FL featuring increased aberrant somatic hypermutation.¹⁵

From genetics to epigenetics and the LME

The high mutation frequency of histone/chromatin-modifying genes underscores a key role for epigenetic dysregulation in the expansion of the premalignant clone and the heterogeneity of FL. The programs modulated by CREBBP, KMT2D, or EZH2, and disrupted in cells carrying these mutations, are implicated in the control of germinal center B-cell state transitions, fate decisions, and cross talk with other immune cells. For instance, these programs govern the expression of major histocompatibility complex class II (MHC-II; a substrate of CREBBP)^6,17,18 and MHC-I (negatively regulated by EZH2).¹⁹ CREBBP-mutated FLs show downregulation of MHC-II, and EZH2 gain-of-function mutations are associated with reduced MHC-I levels.^19,20 These findings suggest that tumor cells may instruct the surrounding LME, and these 2 compartments coevolve over time. For example, FLs developing in compound BCL2-transgenic/EZH2-mutant knockin mice switched from T follicular helper cells to follicular dendritic cell dependence,²¹ and inhibition of EZH2 enhanced response to chimeric antigen receptor (CAR) T cells in a preclinical tFL model by restoring tumor immunogenicity and priming the LME and T-cell cytotoxicity.^22,23 These data suggest that dual targeting of FL and the LME may represent an effective therapeutic approach.

Toward a deeper understanding of the LME

Although the importance of the LME in FL has long been recognized, prior studies relied on limited markers. Advanced multiplexed technologies and multimodal integrative analyses enable detailed characterization of LME complexity, including tumor architecture, spatial organization, and immune-cell interactions, evident across different tumor sites and among individual patients.²⁴ More recently, single-cell analysis of evolutionary dynamics of transformation revealed a shifting LME landscape, with an emerging immune-cell exhaustion signature coevolving with the malignant B-cell phenotype.²⁵ These insights are anticipated to identify novel predictive and prognostic biomarkers, informing precision medicine approaches.

Translating biology into therapy

Although the concept of biomarker-driven precision epigenetic targeting will require clinical validation, preclinical proof of principle has been shown for the preferential sensitivity of CREBBP- and/or EZH2-mutant cells to targeted approaches, including, among others: (1) small-molecule inhibitors of the KAT3 acetyltransferase family, which exploit p300 paralogue lethality and are currently in clinical trials for hematologic malignancies^26,27; and 2) inhibitors of HDAC3, a component of the BCL6 co-repressor complex that opposes CREBBP activity.^6,7 HDAC3i are currently under development and could act as sensitizers when combined with immunotherapy or immunomodulatory agents, as recently shown for EZH2 inhibitors,²² possibly as fixed or intermittent duration to prevent toxicity.

First-line treatment in FL

Outcomes of initial therapy for patients with FL have improved over the last 2 decades, driven by advances in treatment and supportive care.²⁸ The incorporation of CD20-targeted monoclonal antibodies (mAbs) into frontline therapy represents the most significant advancement. Four randomized trials demonstrated that the addition of rituximab to chemotherapy results in improved OS.29, 30, 31, 32 Additional maintenance rituximab for 2 years after CIT induction yields extension in progression-free survival (PFS) but does not improve OS.⁸ Alternatively, some patients, particularly those with lower disease burden, can be managed with rituximab alone.33, 34, 35 Although rituximab remains widely used in clinical practice, the phase 3 GALLIUM study demonstrated that incremental improvement in PFS could be achieved by substituting rituximab with the second-generation CD20 mAb obinutuzumab in combination with chemotherapy.^36,37 Modification of the chemotherapy backbone may also yield improved outcomes. Two randomized trials compared bendamustine-based CIT to CVP (cyclophosphamide, vincristine, and prednisone)– or CHOP (cyclophosphamide, doxorubicin, vincristine, and prednisone)–based treatment and found that bendamustine improved complete response (CR) rate and possibly PFS (with a significant improvement in PFS in 1 trial and numerical improvement in PFS in the other), but did not affect OS.38, 39, 40 In addition, the RELEVANCE trial established lenalidomide plus rituximab as an alternative frontline treatment, with a distinct toxicity profile and similar PFS compared to CIT.⁴¹

With these changes in treatment, newly diagnosed patients with FL can expect to have an initial remission of ≥10 years with CIT and rituximab maintenance and a life expectancy of ≥15 years.^8,28 However, limitations remain for patients with newly diagnosed FL.

CIT is associated with considerable toxicity

Frontline CIT is associated with infections, hematologic side effects, and secondary malignancies. Growing evidence suggests that bendamustine-based therapy may have increased long-term risks compared to other frontline approaches.^40,42 Bendamustine is associated with prolonged lymphopenia and infections. In the phase 3 GALLIUM study, 5% to 6% of patients receiving bendamustine-based treatment had a fatal adverse event, driven primarily by infections.^36,43,44

Improved first-line treatments have not significantly reduced the risk of histologic transformation

Transformation remains a feared outcome for patients with FL and is associated with inferior OS.9, 10, 11 Nonrandomized data suggest that patients receiving rituximab may have modest reduction in this risk,⁴⁵ but transformation is still commonly observed. Although studies have shown high rates of transformation among early progressing patients after bendamustine plus rituximab,⁴⁶ a randomized study comparing rituximab plus CHOP and bendamustine plus rituximab⁴⁰ and well-conducted retrospective studies^47,48 have not shown a significant difference in transformation risk based on chemotherapy backbone.

FL biologic heterogeneity is not factored into treatment decisions

FL is a heterogeneous disease composed of distinct genomic, transcriptomic, and immunologic subtypes that are still being defined.^4,15,49, 50, 51 Outcomes with current frontline therapies are variable, with some patients achieving a functional cure (with ongoing remissions beyond 15 years) and others experiencing primary refractory disease or transformation. Multiple clinical and clinicogenetic risk scores (Follicular Lymphoma International Prognostic Index [FLIPI], FLIPI-2, m7FLIPI, PRIMA-PI, others)^43,52, 53, 54, 55 have been developed to identify patients at low and high risk, but these scores were developed in historical cohorts and do not perform as well for patients receiving current standard treatments.^56,57 Furthermore, they do not guide treatment decisions.

The lack of predictive biomarkers results in both overtreatment and undertreatment of patients with FL

Durable remissions are observed with CIT, but also for some patients receiving less-intensive therapies. A subset of patients (20%-40%) have disease control lasting ≥10 years with rituximab treatment alone.^34,58 Prospective trials have even shown that watchful waiting can be used for ≥10 years for a small subset of patients,⁵⁹ suggesting that some patients require minimal therapy. At that same time, ∼20% of patients will have POD24 after CIT,⁶⁰ including some with transformed disease. Such patients are poorly served by initial treatment and would benefit from alternative frontline treatments.

Frontline treatment for FL may be changing soon

Multiple randomized phase 3 trials are currently enrolling patients with untreated FL (Table 2). Many are testing BsAbs, which have the potential to alter treatment algorithms for FL to a similar degree that rituximab did 20 years ago. Currently, 2 BsAbs targeting CD3×CD20 (mosunetuzumab and epcoritamab) are approved by the US Food and Drug Administration for patients with multiply relapsed FL, although other BsAbs targeting CD3×CD20 (odronextamab) and CD3×CD19 (surovatamig) are in late stages of clinical development. BsAbs have demonstrated high response rates (overall response rate, 78%-96%) and durable remissions (PFS, 15-24 months) in phase 2 trials among patients who had received ≥2 prior lines of therapy.61, 62, 63, 64 In addition, phase 2 trials have demonstrated encouraging results as initial therapy, either as monotherapy or in combination with lenalidomide.65, 66, 67 BsAbs have a distinct side-effect profile compared to CIT, including frequent, mostly low-grade cytokine release syndrome, infections, and immunosuppression. As these drugs are relatively new, predictors of toxicity and long-term safety profiles are still being defined. At least 6 phase 3 trials (with a combined planned enrollment of nearly 5000 participants) are planned or actively enrolling patients. These studies are investigating different strategies, including BsAb monotherapy and combinations with lenalidomide, rituximab, and chemotherapy. These ongoing phase 3 trials may establish new treatment options for patients with untreated FL.

Table 2.

Ongoing phase 3 clinical trials with BsAbs in frontline FL

Clinical trial	N	Description	Patients	Primary end point(s)	Start date/anticipated primary end point date
BsAbs in GELF+ patients
MorningLyte	790	Mosun + Len vs R or O + CHOP or B	Grade 1-3A, FLIPI 2-5	PFS	May 2024/November 2028
EPCORE FL-2	1080	4-way randomization with main comparison between arm A (Epco + Len + R) and arm B (investigator’s choice of R or O + chemotherapy [CHOP or B]) Arm A1: shorter course Epco + Len + R Arm C: Len + R	Stage II-IV, Grade 1-3A	CR30, PFS	February 2024/October 2028
OLYMPIA-1	478	Odro followed by Odro maintenance vs CIT (R-CVP, R-CHOP, or BR, all followed by R maintenance)	Stage III-IV or stage II bulky, Grade 1-3A	CR30	December 2023/April 2029
OLYMPIA-2	733	1:1:1 randomization to R chemotherapy (CVP or CHOP) with R maintenance vs Odro chemotherapy (CVP or CHOP) vs Odro chemotherapy + Odro maintenance	Stage III-IV or stage II bulky, Grade 1-3A	CR30	November 2023/July 2029
SOUNDTRACK-F1	1015	1:1:1 randomization: arms A and B (different schedules of AZD0486 + R) vs arm C (R-CVP, R-CHOP, or BR followed by R maintenance)	Grade 1-3A	PFS	August 2024/April 2031
BsAb in GELF− patients
SWOG2308	600	4 weekly doses of R followed by 5 doses every 8 weeks vs 8 cycles of Mosun	Grade 1-3A, low GELF	PFS	August 2024/March 2032

B, bendamustine; Epco, epcoritamab; Len, lenalidomide; Mosun, mosunetuzumab; O, obinutuzumab; Odro, odronextamab; R, rituximab.

Currently enrolling trials offer key challenges and opportunities. Identifying predictive biomarkers may be more feasible in these frontline trials, given the stark differences in mechanisms of action between treatment arms; however, the differences in design between frontline BsAbs trials (combination partners, duration of treatment, etc) may make it challenging to generalize potential biomarkers. These trials also will not immediately settle questions around the optimal sequencing of these therapies. How will short-term and long-term safety compare between CIT and BsAb-based treatment? How will frontline BsAb-based treatment affect efficacy and safety of subsequent CIT? Will loss of CD20 (observed in a subset of relapses)⁶⁸ have a negative effect on subsequent treatment? If chemotherapy is used in initial therapy, will it impair the efficacy of BsAbs in the relapsed/refractory setting? Will this impact depend upon the chemotherapy used and proximity to BsAb initiation, as observed in patients receiving CAR T cells after bendamustine?⁶⁹ Finally, as we strive to optimize care for patients with FL, we must recognize that large, randomized trials are a key opportunity to better understand how different treatments impact QOL and patient-reported outcomes.

Role of POD24

POD24 was established as a correlate for poor outcomes in FL in 2015, when an analysis of the National LymphoCare study demonstrated marked difference in survival based on remission status at this time point.⁶⁰ This work built on prior reports of inferior survival in patients with short response duration in indolent lymphomas.^70,71 Event-free survival at 12 months after diagnosis and 24 months after treatment are other predictors of outcomes in FL.^72,73 Subsequently, the Follicular Lymphoma Analysis of Surrogacy Hypothesis study analyzed 5225 patients enrolled in 12 multicenter randomized trials receiving frontline treatment and confirmed the impact of POD24 on prognosis in this group of heterogeneously treated patients.⁷⁴ The prognostic impact of POD24 appears less striking among patients undergoing active surveillance compared with those treated with immunotherapy or immunochemotherapy.^72,75, 76, 77

Importance of transformation

Variability in imaging and biopsy practices have contributed to an unclear incidence of transformation within the subset of patients with POD24.^11,46,77, 78, 79, 80 Performing positron emission tomography (PET)–computed tomography scans and biopsies when feasible is of key importance in confirming relapse of FL vs transformed lymphoma, particularly because treatment strategies may be different and are dependent on the frontline treatment received. Before approval of novel treatments, several studies suggested that patients with tFL after CIT benefited from autologous stem cell transplantation (ASCT).^11,81, 82, 83, 84 CAR T cells and BsAbs have improved survival in patients with tFL and,⁸⁴ ASCT is now less frequently deployed.⁸⁵ The timing of transformation is key. Early events affect prognosis most, regardless of whether transformation has occurred. In addition, those with occult transformation at diagnosis represent a distinct biologic group compared with those experiencing later events.

Significance of POD24 on later lines of therapy

The effect of POD24 designation on prognosis in later lines of therapy is unclear. In the relapsed/refractory setting, patients with POD24 and no evidence of transformation have similar response rates to those without POD24 but may have more limited response duration.86, 87, 88 Intriguingly, responses of patients with POD24 receiving CAR T-cell therapy and BsAbs are similar or even higher than those in patients without POD24.^61,62 Unfortunately, these patients can have lower rates of T-cell expansion compared with those without POD24, which could affect the duration of response with CAR T-cell therapy.⁸⁹ Whether these differences are because of prior therapy or biologic aspects of the disease remains unknown.

Future use of POD24

POD24 remains an excellent parameter to predict survival in FL. As patients with POD24 status proceed with additional therapy, they should be re-evaluated at multiple time points to try to better understand biologic differences at each relapse. Analysis of the disease, ideally with material obtained noninvasively, such as ctDNA, should ideally be performed at diagnosis and throughout a patient’s trajectory. A recent retrospective analysis of 141 patients enrolled in the RELEVANCE trial with available serum samples and PET data reported that 6 of 7 patients with detectable ctDNA and a positive PET scan at the end of induction relapsed within 2 years (POD24).⁹⁰ In the future, we aspire to identify patients with high-risk biologic profiles so that we can predict which patient will have POD24. We could then select treatment most likely to be effective in these patients in a tailored approach to avoid early progression.

Clinical trial design

Designing clinical trials in FL presents challenges driven by disease heterogeneity and favorable outcomes with modern therapies. At what point should a patient no longer be managed with watchful waiting and instead be considered a candidate for systemic therapy? Currently, Group d’Etude des Lymphomes Folliculaires (GELF) criteria are used to identify patients in need of treatment, but studies suggest that individual GELF criterion have different prognostic impacts,⁹¹ and the presence or absence of GELF criteria at treatment initiation may not affect PFS with initial treatment.⁹² With remissions to initial therapy exceeding 10 years in some studies, phase 3 trials can take more than a decade to reach a primary end point. What approaches can be used to identify promising therapies more rapidly?

Patient selection

Excellent outcomes with current frontline FL therapies pose a challenge in trial design as they set a high benchmark for therapeutic improvement. Prognostic enrichment, which aims to enroll patients with an increased likelihood of reaching a disease-related end point, can facilitate more rapid completion of clinical studies with smaller sample sizes. Among newly diagnosed patients, identifying those at highest risk remains a challenge.^56,57 Newer prognostic indices, such as FLIPI24, were developed and validated among patients receiving bendamustine-based treatment and appear to be better able to risk-stratify patients in contemporary cohorts.⁹³ Although most frontline FL trials continue to enroll all patients with an indication for treatment, prognostic enrichment is being implemented in some frontline trials; for example, the MorningLyte trial, which restricts enrollment to patients with a FLIPI score of 2 to 5 (Table 2). In the relapsed/refractory setting, POD24 initially appeared to be an attractive candidate for prognostic enrichment; however, trials targeting POD24 patients have been challenging to enroll (eg, SWOG 1608). Because the POD24 population is relatively small and includes many patients with transformed disease, a wider group of high-risk patients with relapsed/refractory FL may need to be targeted in future trials. Enrichment could also be achieved by targeting biologic subgroups of FL most likely to respond to a specific therapy. This approach was used in initial trials of tazemetostat (in which EZH2 mutations were associated with higher response rates but similar PFS),⁹⁴ but has not been widely used in other studies.

Trial end points

Although OS remains the most critical end point, its feasibility as a primary end point in FL trials is limited. PFS has become a well-established primary end point for most phase 3 trials, although it is a poor surrogate for OS.⁹⁵ However, maximizing PFS is not necessarily the primary goal of patients and their physicians. There are numerous examples of treatments that improve PFS but are used selectively or not at all because of toxicity concerns (radioimmunotherapy, ASCT, obinutuzumab, and CD20 mAb maintenance).^{8,37,46,84,96} Safety and tolerability are critical considerations in a disease in which patients have access to multiple highly active treatments. Moreover, a patient’s experience of a treatment regimen is not fully encompassed by traditionally reported adverse events. Patient-reported outcomes and QOL metrics are critical to understand how a patient experiences 1 treatment compared with another, yet there is a dearth of QOL data in FL.⁹⁷ QOL end points should be mandatory for randomized FL trials. Only by integrating both efficacy outcomes and QOL measurements, can we gain a full understanding of the treatment’s impact on patients’ well-being. We acknowledge that current methods of capturing and interpreting QOL data are inadequate, and we advocate for ongoing efforts to improve these tools.

Although PFS is the most common end point in phase 3 trials, it requires extended follow-up, particularly for frontline trials. Waiting for PFS end points prolongs trial duration and can delay availability of new effective treatments. Efforts are underway to develop surrogate end points that can identify a winning treatment regimen more rapidly. CR rate at 30 months (CR30) has been established as a surrogate end point for PFS in first-line FL therapy based on a large analysis of 13 randomized trials of induction and maintenance therapy in FL.⁹⁸ CR30 is being used as a primary or coprimary end point in multiple ongoing frontline phase 3 FL trials (Table 2). Even earlier surrogate end points may be feasible in FL. MRD analyses from multiple phase 3 trials (FOLL5, StiL, GALLIUM, and RELEVANCE) have shown that MRD is associated with PFS among patients receiving frontline CIT.99, 100, 101, 102 These trials assessed MRD using polymerase chain reaction (PCR) to detect a t(14;18) translocation (and in some cases also a clonal immunoglobulin heavy- or light-chain rearrangement). PCR-based MRD testing is only feasible for a subset of patients (50%-77% in these trials), limiting its utility. More modern MRD approaches that use high-throughput sequencing to track dozens or hundreds of tumor reporters have been tested in FL with promising results, albeit in smaller data sets than seen in other hematologic malignancies.^90,103, 104, 105, 106 More data are needed before MRD could be used as a surrogate end point in FL, but there is a precedent, as the US Food and Drug Administration and European Medicines Agency now accept MRD as a surrogate end point for accelerated approval of treatments for multiple myeloma.

MRD could also be used as a tool to adapt therapy for patients with FL. The ability to selectively escalate or de-escalate therapy for FL could represent a major advance for patients with FL, but this approach has potential pitfalls. The FOLL12 trial randomly assigned patients with treatment-naive FL to receive standard rituximab maintenance or MRD-adapted maintenance therapy after frontline CIT. In the MRD-adapted arm (which used PCR-based MRD testing), MRD-negative patients received no maintenance and MRD-positive patients received more intensive maintenance with the goal of achieving MRD negativity. The study reported inferior PFS in the MRD-guided arm, sounding a note of caution for MRD-guided treatment strategies in FL.¹⁰⁷

Cure

FL has historically been considered an incurable disease. With more effective therapies, we advocate for a change in terminology.⁸ The FLIPI24 prognostic model found that 40% of patients have only a 3% risk of dying from lymphoma within 10 years of their diagnosis.¹⁰⁸ An older patient in remission for 12 years after frontline therapy who subsequently dies of another cause may not have technically been cured but has achieved a functional cure.¹⁰⁹ Such patients are at higher risk of treatment-related toxicities than of death from lymphoma, and clinicians should be cautious not to overtreat them. In these scenarios, QOL should be emphasized. We advocate for fewer monitoring tests in asymptomatic patients to avoid procedures and treatment that are unlikely to extend OS. Models of treatment-related mortality are being developed in patients with FL receiving second- and third-line therapies. A goal is to identify those with relapsed disease who can be monitored without treatment and eventually extrapolate these findings to the frontline setting. Such models would aid clinicians and patients as they face treatment decision-making throughout their disease course.

In other patients, we may be able to refine the definition of cure and potentially achieve it with novel therapeutic approaches. Young, fit patients are among those who may be considered for curative strategies. Patients with symptomatic, high tumor burden disease could be studied in clinical trials to assess whether aggressive novel treatments, such as CAR T cells, are superior to standard CIT. We could even consider whether young patients with low burden disease should be studied with similarly aggressive treatment vs active surveillance. With these novel therapies, long-term follow-up is essential. Clinical trials frequently do not routinely monitor beyond 8 to 10 years, and therefore later events are not captured. We emphasize the importance of extended follow-up in clinical trials enrolling patients with FL, especially in the CAR T-cell therapy, in which long-term toxicity is a concern. We advocate for pharmaceutical companies and other funding sources to include expenses related to long-term follow-up in their budgets. The use of telemedicine and virtual applications in this setting could improve the patient burden and potentially decrease cost. It would also be informative to study the changes in molecular characteristics of patients treated with aggressive up-front strategies compared with standard treatment or active surveillance, to better understand how these interventions affect disease trajectory.

Actualizing cure

Achieving curative strategies for FL will require a comprehensive molecular understanding of the mechanisms driving its heterogeneity, early progression, and transformation. Outstanding requirements include a functional analysis of mutations targeting the noncoding genome, which represents 98% of the cell’s genetic material yet remains almost completely unexplored; the development of preclinical models that accurately recapitulate the tumor and its microenvironment, such as recent syngeneic mouse models of tFL and patient-derived xenografts²²; and, ideally, the construction of scalable organoid systems for drug testing and mechanistic studies.110, 111, 112 Equally important is the establishment of large, harmonized data sets from uniformly treated patients, particularly those enrolled in modern clinical trials and those treated in real-world clinical practice, to validate potential biomarkers for risk stratification and treatment guidance. Addressing these challenges will require a concerted effort from investigators with crossdisciplinary expertise across genomics, immunology, epigenetics, and computational biology. The role of private philanthropic funds, such as the one sponsoring this workshop, will be instrumental in achieving this goal.

Ensuring access

Finally, we must expand access to novel therapies for patients with FL. Many from around the world are unable to benefit from the approaches reviewed at this workshop. Patients from diverse backgrounds, countries, and settings should be enrolled on clinical trials in FL and included in real-world studies. Widespread global education efforts should be made to patients and hematologists/oncologists to gain comfort in treatment algorithms and management of side effects of novel therapies. Patient focus groups and discrete choice experiments should be conducted in FL to better understand patient preferences and engage them in clinical trials.^113,114 These strategies could facilitate broader uptake of these treatments and improved outcomes.

Conclusions

The 2024 International Follicular Lymphoma Scientific Workshop comprehensively reviewed areas of opportunity and unmet needs in FL. This provided a platform for lymphoma experts to discuss the latest treatments and develop strategies for optimizing patient care and clinical trial design. Continuing efforts are needed to better understand the underlying biology of this heterogenous disease and to promote collaborative efforts to advance biomarkers, disease models, and other resources. Integrating compassionate care and patient engagement into clinical trials is essential. Our strategy must involve not only focusing on eradication of the disease but also enhancing QOL. Including patient voices and experiences in research can drive more patient-centric approaches and improve the overall treatment landscape for FL.

Conflict-of-interest disclosure: R.M reports serving on the advisory board for Genmab, Bristol Myers Squibb (BMS), AbbVie, Ipsen, and Kite; consulting for DG Medicine; receiving honoraria from Genmab and AbbVie; receiving institutional research funding from Merck, BMS, Genmab, and Genentech/Roche. S.C.R. reports consulting for AbbVie, ADC Therapeutics, BMS, Genmab, Incyte, Ipsen, Karyopharm, Kite, and Pfizer; serving on the data and safety monitoring board of Karyopharm; and receiving research funding from Constellation, Genentech/Roche, Ipsen, and Karyopharm. P.A. is a consultant for Merck, BMS/Celgene, Pfizer, Affimed, Adaptive, Infinity, ADC Therapeutics, MorphoSys, Daiichi Sankyo, Miltenyi, Tessa, Genmab, C4 Therapeutics, Enterome, Regeneron, Epizyme, AstraZeneca, Genentech/Roche, Xencor, Foresight Diagnostics, ATB Therapeutics, and Mabqi; reports research funding from Kite, Merck, BMS/Celgene, Affimed, Adaptive, Tensha, Otsuka, Sigma Tau, Genentech/Roche, IGM Biosciences, and AstraZeneca; and reports honoraria from Merck and BMS. J.P.L. has received research support from National Institutes of Health/National Cancer Institute, Leukemia and Lymphoma Society, Genentech Foundation, Lymphoma Research Foundation, Follicular Lymphoma Foundation, Epizyme, and Janssen; consulting fees from AstraZeneca, BeiGene, Caribou Biosciences, Foresight, Genentech, Kyowa Kirin, Novartis, Ono, Pfizer, Regeneron, Sail Bio, and Treeline Biosciences. L.N. has received honoraria for consulting from AbbVie, AstraZeneca, BMS, Genentech, Genmab, Gilead/Kite, Incyte, Ipsen, Janssen, Novartis, Regeneron, and Takeda; research support from BMS, Daiichi Sankyo, Genentech, Genmab, Gilead/Kite, Janssen, Merck, Novartis, and Takeda. S.M.S. has received consulting fees (time limited and/or one time) for Genmab, Regeneron, and Foresight Diagnostics; and her spouse is employed by Caris Life Sciences. A.D.Z. has served as a consultant (including expert testimony) for, and has received honoraria from, Genentech, Ipsen, AbbVie, Allogene, Curio Science, DAVA Oncology, BeOne, Ono Pharma, Kite, AstraZeneca, Genmab, and Eli Lilly; Has received research funding from Genentech/Roche, Arvinas, BeOne, GSK, and Pharmacyclics (Rollover Study); serves as a data monitoring committee member for BMS/Celgene and Juno. C.C. has received honoraria from Incyte, AbbVie, Genentech/Roche, and Genmab; has served as a consultant for Incyte, AbbVie, Genentech/Roche, and Genmab; and has received research funding from Genmab, Gilead , and Genentech. J.C. has received research funding (to Mayo Clinic) from Genentech, Genmab, and BMS; and serves on the science advisory board for Protagonist. M. Green has received research funding from Sanofi, Kite/Gilead, AbbVie, and Allogene; consulting fees from AbbVie and Allogene; serves on the advisory board for BMS, Arvinas, and Johnson & Johnson; has received honoraria from BMS, Daiichi Sankyo, and DAVA Oncology; and has stock ownership in Melbridge Therapeutics and Shenandoah Therapeutics. B.K. reports consulting for AbbVie, BMS, GSK, Roche, BeiGene, Eli Lilly, ADC Therapeutics, AstraZeneca, Genentech, Pfizer, Merck, and Incyte; and receiving research funding from AstraZeneca, BeiGene, and Roche. R.K. has received research funding (to institution) from AbbVie, AstraZeneca, BMS, and Roche. M.J.M. has received research funding from Roche/Genentech and BMS; and honoraria from MD Education. B.N. has received grant/research support from BMS; and honoraria from BeiGene, Diatech Pharmaceutical, and Mabqi. A.J.R. is an employee of Leica Microsystems. E.L. has received consulting fees from Pfizer and ADC Therapeutics. G.S. is a member of advisory boards, consulting committees, or data monitoring committees for AbbVie, BeiGene, BMS, Canopy, Daiichi Sankyo, Ellipses, Genentech/Roche, Genmab, Janssen, Incyte, Ipsen Kite/Gilead, Eli Lilly, Merck, Novartis, and SERB Pharmaceuticals; and has received research support from AbbVie, Genentech, Genmab Janssen, and Ipsen, which was managed by his institution. L.S. has received consulting fees or honoraria from AbbVie, AstraZeneca, BeiGene, Cargo, Chugai, BMS, Eli Lilly, Genmab, Kite/Gilead, Incyte, Janssen, Merck, Seagen, and Roche/Genentech; and research funding from Roche/Genentech. L.P. has received research grant from AstraZeneca. A.S.L. has received consulting fees from Genmab, Kite, and Pierre Fabre. The remaining authors declare no competing financial interests.