FormalPara Key Points

Personalized treatment strategies for patients with relapsed or refractory follicular lymphoma require thoughtful consideration of prior therapies and patient-related factors to identify those who are likely to benefit from novel treatment options.

Chimeric antigen receptor T-cell therapies have emerged as a standard of care in the third-line setting providing durable remission and manageable safety for patients with relapsed or refractory follicular lymphoma.

Optimal sequencing of therapies for patients with relapsed or refractory follicular lymphoma remains to be established based on real-world data and future clinical trials.

1 Background

Follicular lymphoma (FL) is the second most common subtype of B-cell non-Hodgkin lymphoma (NHL) in the USA, with an incidence rate of 2.6 per 100,000 persons per year based on data collected from 2015 to 2019 [1, 2]. Follicular lymphoma represents about 20% of all NHL cases [2] and typically occurs in patients aged 50–60 years; however, because of its indolent nature, FL is often not diagnosed until patients are in their mid-60s [3]. Follicular lymphoma has a lengthy disease course with alternating periods of remission and relapses that are associated with a risk of transformation and development of secondary malignancies [4,5,6]. Currently, there is no clear standard of care or preferred treatment approach for patients with relapsed or refractory (r/r) FL. The purpose of this review is to provide evidence-based guidance for the treatment of r/r FL focusing on chimeric antigen receptor (CAR)-T cell therapies.

2 Management

2.1 Current Treatment Landscape

Follicular lymphoma remains an incurable disease [7]. Approximately 20% of the patients experience relapse within 2 years after first-line treatment [8]. Patients with r/r FL typically have worsening treatment outcomes with increasing lines of therapy [9]. Patients who experience early disease progression (prior to 2 years post front-line therapy) are at an increased risk [8]. The National Comprehensive Cancer Network [10] and European Society for Medical Oncology [11] clinical practice guidelines suggest several treatment approaches for patients with r/r FL; however, the treatment guidelines do not provide a consensus on a preferred treatment sequence or approach. Common treatment options for patients with r/r FL in second-line treatment are similar to first-line therapies and include rituximab or lenalidomide plus rituximab (R2); bendamustine plus obinutuzumab or rituximab plus cyclophosphamide, doxorubicin, vincristine, and prednisolone (R-CHOP); rituximab plus cyclophosphamide, vincristine, and prednisolone (R-CVP); and other chemotherapy-based regimens.

During the course of their disease, patients with FL often receive multiple treatment regimens with a natural history that can span years [9, 12,13,14]. Expected efficacy outcomes, such as overall response rate (ORR), progression-free survival (PFS), and overall survival (OS), are often diminished with each line of therapy [9]. Although hematopoietic stem cell transplant is an option for patients with relapsed FL, the optimal timing is not well defined [9] and < 10% of patients will receive a hematopoietic stem cell transplant in the second-line setting and beyond in the USA [13]. In the modern era of expanding treatment options, more data are needed to better understand how to most effectively sequence available therapies after relapse in patients with FL to improve remission durability.

3 Approved and Investigational Third-Line Therapies for r/r FL

The therapeutic landscape for r/r FL is rapidly evolving. Third-line and subsequent treatment options include second-line therapies not previously used; the enhancer of zeste homolog (EZH) 2 inhibitor tazemetostat for patients with an EZH2 mutation after 2 previous therapies or EZH2 wild-type or unknown in patients who have no satisfactory alternative treatment options, recently approved anti-CD19 CAR-T cell therapies such as axicabtagene ciloleucel and tisagenlecleucel after two or more prior systemic therapies, and bispecific antibodies (BsAbs). Clinical trial data supporting approved and investigational therapies for patients with r/r FL are presented in Tables 1 [15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35] and 2 [3, 36,37,38,39,40,41,42,43,44,45,46,47,48,49].

Table 1 Approved and investigational treatments for patients with r/r FL
Full size table
Table 2 Bispecific antibodies and CAR-T cell therapies for the treatment of patients with r/r FL
Full size table

3.1 Monoclonal Antibodies

The treatment paradigm for patients with B-cell NHL has been transformed with the introduction of monoclonal antibodies targeting CD20 [50], such as rituximab and obinutuzumab (both approved by the US Food and Drug Administration [FDA]) [18, 51]. For patients with rituximab-refractory disease, combined therapy with obinutuzumab has also been approved [18]. Obinutuzumab-based treatment (obinutuzumab plus bendamustine) has a reported ORR of 68% and substantial OS and PFS benefits (median OS not reached; median PFS was 25.3 months after median follow-up of 32 months) in patients with rituximab-refractory r/r FL [19]. Other strategies being explored in the r/r FL setting include monoclonal antibodies that target different epitopes on CD20, a modified fragment crystallizable portion of the antibody with enhanced cell-mediated cytotoxicity, and antibodies that target other B-cell surface antigens such as CD19, CD22, and CD79. For example, tafasitamab, a monoclonal antibody that targets CD19, is currently being evaluated in combination with R2 in the InMIND phase III trial [20].

Overall, data from clinical trials in patients with r/r FL have demonstrated that monoclonal antibodies targeting B-cell surface antigens combined with lenalidomide or chemotherapy can achieve ORRs that range from 56% [52] to 79% [26]. Median PFS ranged from ~ 25 months (obinutuzumab plus bendamustine ± rituximab) [19, 53] to 50.5 months (R2) [54]. Although well tolerated in most patients, key safety concerns associated with monoclonal antibodies include infusion-related reactions, neutropenia, thrombocytopenia, pyrexia, diarrhea, headache, infections, chills, and insomnia. A recent meta-analysis of second-generation CD20 monoclonal antibodies suggests a higher incidence of adverse events (AEs) and serious AEs for these monoclonal antibodies compared with rituximab [55].

3.2 Kinase Inhibitors

Several PI3K inhibitors (copanlisib, umbralisib, duvelisib, and idelalisib) that showed promising efficacy and gained FDA approval have been recently withdrawn from the market by the drug manufacturers because of a variety of challenges associated with the data analysis of ongoing trials, trial enrollment, market influences, and safety concerns [56,57,58,59,60]. The FDA discouraged the accelerated approval of another PI3K inhibitor, zandelisib, recommending evaluation in a randomized trial rather than a single-arm study [27]. Many of the PI3K inhibitors have been associated with rare and often unpredictable serious immune-related AEs (e.g., diarrhea/colitis, transaminitis, and pneumonitis) [61] and immune suppression, limiting their effective use in some patients with r/r FL [62,63,64].

Other kinase inhibitors have been evaluated for efficacy and safety in patients with r/r FL. In a single-agent, phase II trial, ibrutinib, a small-molecule inhibitor of Bruton’s tyrosine kinase, demonstrated modest activity with an ORR of 20.9% after a median follow-up of 27.7 months, although it failed to meet the primary endpoint [30]. Zanubrutinib, a more selective next-generation Bruton’s tyrosine kinase inhibitor with fewer off-target effects, showed superior efficacy outcomes when added to obinutuzumab as compared with obinutuzumab alone (ORR 69% vs 46%; complete response rate [CRR] 39% vs 19%; 18-month duration of response [DOR] 69% vs 42%; and median PFS 28.0 months vs 10.4 months) in patients with r/r FL with similar AE profiles between groups [31]. In March 2024, zanubrutinib received accelerated approval by the FDA for the treatment of r/r FL in combination with obinutuzumab after two or more lines of systemic therapy on the basis of response rate and response durability [35]. A confirmatory trial will be required to verify these findings and indication approval [35].

3.3 EZH2 Inhibitors

Approximately 20–25% of patients with FL have EZH2-mutation positive disease [57, 65]. Tazemetostat, a selective EZH2 inhibitor, gained accelerated FDA approval for the treatment of r/r FL in 2020 [24]. In the phase II registration study, efficacy outcomes for tazemetostat in the EZH2 mutant versus EZH2 wild-type cohorts were ORR of 69% vs 35%, median DOR of 10.9 months versus 13 months, and median PFS of 13.8 months versus 11.1 months, respectively [23]. Because tazemetostat demonstrated efficacy regardless of patient EZH2 mutational status, albeit higher efficacy in patients with EZH2 mutation-positive disease, FDA approval was granted regardless of mutational status. Tazemetostat was well tolerated in both arms with low rates of grade ≥ 3 treatment-related AEs, including thrombocytopenia (3%), neutropenia (3%), and anemia (2%), and no treatment-related deaths [23]. Confirmatory phase III studies in relapsed indolent lymphoma are ongoing [66].

3.4 Allogeneic and Autologous Stem Cell Transplant

A hematopoietic stem cell transplant following high-dose chemotherapy remains a common treatment approach in patients with r/r FL to try to achieve prolonged remission after an initial or subsequent relapse. Two types of stem cell transplant (SCT) are available to patients with r/r FL: allogeneic (allo) SCT and autologous (auto) SCT [67]. While alloSCT eliminates the risk of potential lymphoma cell contamination, the elevated risk of graft-versus-host disease and increased treatment-related mortality has resulted in a bias toward autoSCT [67, 68]. Development of nonmyeloablative regimens over the last few decades, however, has reduced treatment-related mortality and increased patient access to alloSCT [67, 68]. A recent analysis comparing outcomes in patients with r/r FL who received either allo-SCT or autoSCT demonstrated significantly better OS in patients who received alloSCT compared with patients who received autoSCT (62% vs 46%; p = 0.048). Similar trends that favored alloSCT were observed for PFS (52% vs 31%; p < 0.001) and 8-year relapse rates (11% vs 43%; p < 0.0001). Non-relapse-related mortality was similar between the two groups (15% vs 11%) [67]. Both alloSCT and autoSCT can place a physical challenge on the patient; physicians should assess a patient’s frailty and functionality prior to proceeding.

3.5 Checkpoint Inhibitors

Despite FL’s dependence on microenvironment signaling, immune checkpoint inhibitors (nivolumab and pembrolizumab) have shown limited activity as monotherapy in patients with r/r FL [32, 33]. Pembrolizumab monotherapy in patients with r/r low-grade FL showed a median PFS of 3.4 months and a median OS was not reached after a median follow-up of 6.5 months [33]. Overall, the most common grade 3/4 treatment-related AEs with pembrolizumab in patients with FL were neutropenia, thrombocytopenia, anemia, and dyspnea [33].

3.6 Antibody-Drug Conjugates

Polatuzumab vedotin, an anti-CD79b monoclonal antibody conjugated to the microtubule-disrupting agent monomethyl auristatin E, demonstrated modest efficacy (47% ORR) in a phase I study in a subset of patients with r/r indolent NHL [69]. A phase II trial of rituximab and bendamustine with or without polatuzumab vedotin found that CRRs at 6–8 weeks were similar in both treatment arms in the r/r FL subgroup [34]. Polatuzumab vedotin in combination with lenalidomide, obinutuzumab, venetoclax, and atezolizumab is also being evaluated in patients with r/r FL [70,71,72].

3.7 BsAbs

Bispecific antibodies are an emerging class of drugs with novel structures that target CD3 and tumor-specific antigens simultaneously to enhance cytotoxicity of T cells. This class of drugs has the potential to be manufactured on a large scale with minimal interpatient variability.

The most notable AEs associated with BsAbs have been similar to other cellular therapies and include cytokine release syndrome (CRS) and neurotoxicity, including immune effector cell-associated neurotoxicity syndrome (ICANS) [46, 73] (Table 2 [3, 36,37,38,39,40,41,42,43,44,45,46,47,48,49]). In addition, serious infections (febrile neutropenia, pneumonia, upper respiratory infections), chronic infections (including viral infections), and sepsis, which can lead to life-threatening or fatal events, have been reported [37]. Tumor flare has been noted following an infusion, manifesting as pleural effusions and the enlargement of areas affected by lymphoma [46, 73, 74]. As such, caution and close monitoring for AEs are recommended in high-risk patients with risk factors such as high tumor burden, circulating disease, increased lactate dehydrogenase, impaired organ function, or a history of recurring or chronic infections [73].

Blinatumomab, a CD19/CD3 BsAb, showed promising results with an ORR of 80% in the phase I trial that included 28 patients with r/r FL [42] (Table 2 [3, 36,37,38,39,40,41,42,43,44,45,46,47,48,49]). Adverse events included infections and neurological events (NEs) in 50% (grade ≥ 3, 11%) and 71% (grade 3, 22%) of patients, respectively. The need for a continuous blinatumomab infusion, however, poses some challenges and may negatively impact a patient’s quality of life [42, 75]. Mosunetuzumab, a CD20/CD3 BsAb, has shown clinical benefits in patients with r/r NHL, even after CAR-T cell therapy [76]. In the pivotal phase II study, mosunetuzumab demonstrated durable efficacy (ORR 78%; CRR 60%; median DOR and median PFS were not reached after a median follow-up of 27 months) and a manageable safety profile (CRS 44%, predominantly grade 1/2 events) [36, 77]. Neurological events consistent with ICANS observed following treatment with mosunetuzumab included a confusional state (3%; grade 1/2), disturbance in attention (1%; grade 1), and cognitive disorder (1%; grade 1) [36]. The FDA recently approved mosunetuzumab for the treatment of adult patients with r/r FL after two or more lines of prior systemic therapy [37]. Glofitamab, another T-cell-engaging CD20/CD3 BsAb, induced high response rates in patients with r/r FL, both as a monotherapy (ORR 81%; complete metabolic response 70%) and in combination with obinutuzumab (ORR 100%; complete metabolic response 74%) [78]. Myelosuppression was more common in patients who received glofitamab in combination with obinutuzumab compared with glofitamab alone (neutropenia 58% vs 26%; anemia 37% vs 25%; thrombocytopenia 32% vs 11%). Cytokine release syndrome rates following glofitamab treatment were also elevated when administered in combination with obinutuzumab (79%) compared with glofitamab as monotherapy (66%) [78]. No patient in either the monotherapy or combination cohort experienced ICANS [78]. Patients with r/r FL treated with epcoritamab, a CD20/CD3 BsAb, achieved an ORR of 90% [45]. There were no grade 3 or higher CRS events with epcoritamab monotherapy (grade 1/2, 59%). Neurological events were experienced by 6% of patients (grade 1, 3%; grade 3, 3%) [45]. Combination treatment of epcoritamab with R2 in patients with r/r FL (27 patients) showed a 100% ORR and a 93% complete metabolic response [79]. Cytokine release syndrome was observed in 50% of patients, with most cases being grade 1/2 and occurring in the first cycle of treatment. One patient experienced grade 2 ICANS [79]. The benefit of epcoritamab with R2 was still observed in patients with high-risk disease, including those with progression of disease within 2 years of initial chemotherapy (POD24) [80]. Odronextamab, a hinge-stabilized, fully human, IgG4-based CD20/CD3 BsAb, showed a 91% ORR and 72% CRR, with a median DOR of 15.8 months [46]. Cytokine release syndrome was reported in 61% of patients with r/r NHL, with grade 3 and 4 CRS reported in 6% and 1% of patients, respectively. Immune effector cell-associated neurotoxicity syndrome-like events were reported in 12% (any grade) and 3% (grade ≥ 3) of patients. Grade ≥ 3 ICANS-like events included confusional state, somnolence, encephalopathy, aphasia, and cognitive disorder [46] (Table 2 [3, 36,37,38,39,40,41,42,43,44,45,46,47,48,49]).

4 CAR-T Cell Therapy in r/r FL

Chimeric antigen receptor T-cell therapies that target CD19 have demonstrated durable responses with an acceptable safety profile in the treatment of patients with r/r FL, including those with high-risk disease characteristics (higher Follicular Lymphoma International Prognostic Index score, high tumor burden, POD24, or double refractory) [3, 40]. Two CAR-T cell therapies, axicabtagene ciloleucel and tisagenlecleucel, have gained FDA approval for the treatment of adult patients with r/r FL [38, 41]. The FDA and European Medicines Agency granted approval for axicabtagene ciloleucel for the treatment of adult patients with r/r FL after two or more (FDA) and three or more (European Medicines Agency) lines of systemic therapy, including the combination of an anti-CD20 monoclonal antibody and an alkylating agent [38, 39]. In the ZUMA-5 registrational phase II trial, axicabtagene ciloleucel elicited an ORR of 94% and CRR of 79% in patients with r/r FL [3] (Table 2 [3, 36,37,38,39,40,41,42,43,44,45,46,47,48,49]). Any grade and grade ≥ 3 CRS were reported in 78% and 6% (one patient had grade 5 CRS) of the 124 patients with FL, whereas any grade and grade ≥ 3 NEs were reported in 56% and 15%, respectively [3]. Three-year ORR and CRR were supportive of previous data cuts; median DOR was 38.6 months and median PFS was not reached for patients without POD24 and was 40.2 months for patients with POD24 [81]. Tisagenlecleucel received accelerated FDA approval for the treatment of adult patients with r/r FL after two or more lines of systemic therapy (including a CD20 antibody and an alkylating agent) or relapsed after autoSCT based on the ELARA trial [41] (Table 2 [3, 36,37,38,39,40,41,42,43,44,45,46,47,48,49]). In 94 patients, the ORR was 86.2% and the CRR was 69.1% [40]. Median DOR was not reached, with an estimated 9-month DOR rate of 86.5% in patients who achieved a complete response [40]. Adverse events occurring within 8 weeks of infusion were CRS 48.5% (grade ≥ 3, none), NEs 37.1% (grade ≥ 3, 3%), and ICANS 4.1% (grade ≥ 3, 1%), with no treatment-related deaths [40].

A long-term follow-up has suggested that CAR-T cell therapy in patients with B-cell malignancies can induce long-term remissions with a rare occurrence of late AEs [82]. A 5-year extended follow-up study reported that most of the clinical responses to tisagenlecleucel in the first year were sustained after 5 years, including among patients with r/r FL [83]. Additionally, while data for approved CAR-T cell products in r/r FL are limited, real-world reports for diffuse large B-cell lymphoma show patients aged ≥ 75 years experienced similar efficacy and rates of CRS and ICANS compared with patients aged < 75 years [84]. These findings, should they translate to FL, suggest high response rates to CAR-T cell therapy can be achieved in older adult patients, with minimal safety concerns.

TRANSCEND FL is an ongoing phase II, open-label, multicenter study currently evaluating the efficacy and safety of lisocabtagene maraleucel, another CD19-directed CAR-T cell therapy, in adult patients with r/r FL or marginal zone lymphoma [47] (Table 2 [3, 36,37,38,39,40,41,42,43,44,45,46,47,48,49]). With a median follow-up of 18.1 months, the primary analyses demonstrated an ORR and CRR of 95.7%. Duration of response and PFS were not reached; 12-month DOR and PFS were 89.8% and 91.3%, respectively. Cytokine release syndrome occurred in 52% of patients (grade ≥ 3, 0%), while NEs occurred in 17% (grade 3, 4%). One treatment-emergent AE-related death occurred, in the context of disease progression, due to grade 5 macrophage activation syndrome [48].

4.1 Logistics of CAR-T Cell Therapy

Although promising efficacy has been reported in patients with relapsed disease, the integration of CAR-T cell therapy into the current treatment paradigm for patients with r/r FL requires careful consideration of many factors. First, to improve T-cell collection and to optimize the production of CAR-T cells, appropriate washout periods for previous treatments are recommended when feasible. Recent findings suggest some therapeutics, such as bendamustine, may continue to have a negative impact on CAR-T clinical outcomes months after treatment is stopped and are best avoided completely if clinically feasible. In patients who have received bendamustine, despite the 40-min half-life, the optimal washout prior to leukapheresis may be 9 months or longer to avoid deleterious effects on T-cell fitness [85, 86]. The European Society for Blood and Marrow Transplantation guidelines suggest that patients who have undergone prior alloSCT should not receive immunosuppressants and should be free from graft-vs-host disease for at least 1 month before undergoing leukapheresis for CAR-T cell therapy [87]. The American Society for Transplantation and Cellular Therapy guidelines also indicate that comorbidities such as diminished organ function may affect treatment decisions [88]. Next, appropriate patient selection for CAR-T cell therapy requires a thorough understanding of not only prior lines of therapy (more than two systemic therapies), but also other associated aspects such as patient-related factors (disease status, age, tumor burden, comorbidities), insurance coverage, caregiver support, and access to treatment center (Fig. 1) [89,90,91]. Typically, caregiver support is recommended for at least 30 days from the time of CAR-T cell infusion, with additional support for travel arrangements for up to 60 days afterward. Finally, the choice between CAR-T cell products should be made by weighing several factors, including efficacy, toxicity profiles, risk of AEs due to prior treatments, manufacturing times, and other logistical aspects. Differences in the incidence of potentially life-threatening AEs such as CRS and NEs should also be included in the choice of CAR-T cell therapy.

Fig. 1
figure 1

Logistics of chimeric antigen receptor (CAR) T-cell therapy for patients with relapsed or refractory follicular lymphoma (r/r FL). Key steps and considerations for patients undergoing CAR-T cell therapy for r/r FL. Caregiver support varies by treatment center. AEs adverse events, CRS cytokine release syndrome

Full size image

4.2 Bridging Therapy, Lymphodepletion Chemotherapy, and CAR-T Cell Infusion

Many patients receive bridging therapy to maintain disease control or control symptoms while CAR-T cells are being manufactured (Fig. 1) [90]. Lymphodepleting chemotherapy may be administered at the discretion of the physician based on the patient’s disease status. Physicians may consider omitting lymphodepleting chemotherapy in patients who are cytopenic 1 week before the planned CAR-T cell infusion [41]. Typically, lymphodepletion therapy is concluded 2–14 days before infusion of CAR-T cells and has been shown to prolong the persistence of infused cells enabling sustained responses [38, 41, 90]. Despite the negative impact of bendamustine on CAR-T outcomes when used prior to leukapheresis, administration as a lymphodepleting regimen prior to infusion was both safe and effective with fewer AEs and outcomes similar to fludarabine/cyclophosphamide [85, 92, 93]. Chimeric antigen receptor T cells are administered as a single intravenous infusion in the inpatient or outpatient setting depending on the CAR-T cell therapy being administered, the patient’s risk factors for severe CRS, and the treating center’s capabilities to treat outpatients [94]. While not common, CAR-T manufacturing failures (i.e., inability to produce product or a final product that does not meet release criteria) do occur for a subset of patients, which may result in a significant delay of treatment. Although manufacturing success rates have been continually improving [95], contingency planning for alternative therapy options or re-manufacturing of CAR-T cells is a good practice to mitigate any unintended treatment interruptions.

4.3 Post-Infusion Monitoring and Management of AEs in the Short and Long Term

After infusion, patients should be monitored for signs of CAR-T cell therapy-associated AEs and managed accordingly (Fig. 1). High-risk patients may need to be monitored as inpatients, whereas patients who undergo infusion in the outpatient setting are often managed by outpatient teams and trained caregivers. In addition, it is recommended that patients refrain from driving for 8 weeks after CAR-T cell infusion and remain in close proximity to the treatment center for at least 4 weeks. Subsequently, patients are monitored for late-onset or chronic AEs during regularly scheduled follow-up visits with coordination between the treatment center and local oncologist, caregivers, and support teams. Other common AEs, such as cytopenia and infections, can occur following discharge. Practitioners are encouraged to regularly follow up on patients’ status with blood count monitoring, surveillance for recurrent infections, and assessment of immunoglobulin levels. Among patients with r/r FL enrolled in the ELARA clinical trial, prolonged grade ≥ 3 cytopenia was reported in 20% of patients and 8% had cytopenia at 12 months following tisagenlecleucel infusion [3, 40]. Similarly, 33% of patients with r/r FL in the ZUMA-5 trial reported prolonged grade ≥ 3 cytopenia ≥ 30 days after an axicabtagene ciloleucel infusion [96]. Management of cytopenia may include growth factor support as needed, prophylactic antibiotics, and/or transfusion. The frequency of follow-up care depends on several factors, including institutional preference and experience.

Other AEs that have long-term effects are B-cell aplasia that leads to hypogammaglobulinemia and the risk of infection among patients with prolonged cytopenia [38, 41, 97, 98]. Careful benefit-risk evaluation and clinical judgment are required to assess the use of CAR-T cell therapy in patients with r/r FL. High tumor burden and previous therapies may place patients at an elevated risk of CAR-T-associated AEs, especially CRS [90, 97, 98].

5 Practical Guidance for CAR-T Cell Therapy in Patients with r/r FL

Based on the available clinical trial data and the authors’ experience treating patients with r/r FL, the following are our recommendations for initiating CAR-T cell therapy in this patient population (Fig. 2). Patients who have received two or more prior therapies for r/r FL should undergo a rebiopsy to determine if there has been a histological transformation, especially among patients with more aggressive symptomology (e.g., drenching night sweats, rapid lactate dehydrogenase increase, rapid increase in tumor mass). Patients with transformed FL should receive individualized therapy according to large B-cell lymphoma treatment guidelines considering the types of previous therapies and comorbidities. For patients whose disease relapses without transformation, a risk assessment is recommended to help determine the next course of therapy. Patients with prolonged remissions prior to disease progression can be considered non-high risk and may receive individualized therapy, including second-line therapies that were not previously used, BsAbs, zanubrutinib and obinutuzumab, anti-CD20 monoclonal antibodies, or tazemetostat for patients with EZH+ disease or in patients without other treatment options. Following relapse to one of these treatments, a patient’s eligibility for CAR-T therapy should be re-assessed. Enrollment in a rationally designed clinical trial can always be considered based on a patient’s needs and/or physician discretion; however, clinical trial enrollment should also be strongly considered for patients whose disease relapses following currently approved therapies, including CAR-T.

Fig. 2
figure 2

Practical guidance for chimeric antigen receptor (CAR) T-cell therapy in patients with with relapsed or refractory follicular lymphoma (r/r FL). Recommended strategy for identifying patients with r/r FL who are appropriate for receiving CAR-T cell therapy. These recommendations are based on the results from clinical trials of CAR-T cell therapy in patients with r/r FL and the authors’ clinical experience treating patients with r/r FL. BsAb bispecific antibody, CD cluster of differentiation, DLBCL diffuse large-B-cell lymphoma, ISRT involved-site radiation therapy, mAbs monoclonal antibodies, POD24 progression of disease within 2 years of initial chemotherapy, SCT stem cell transplant

Full size image

Patient identification for CAR-T cell therapy should be guided by an individualized approach that weighs the benefits and risks of therapy based on patient and disease characteristics and the clinical experience of the treating physician. The patient’s ability to withstand the potential treatment-related AEs is an important consideration in determining the eligibility for CAR-T cell therapy. In general, patients with adequate organ function (cardiac, pulmonary, renal, hepatic, and hematologic), Eastern Cooperative Oncology Group score of 0–1, and no active infection are potential candidates for CAR-T cell therapy [99]. The pace at which the patient’s disease is progressing can also influence treatment decisions because of the logistical hurdles associated with CAR-T cell therapy. For patients with rapid disease progression, it is important to be mindful of the time needed to move through insurance approval, leukapheresis collection, manufacturing, bridging therapy, lymphodepletion, and infusion. Patients who are ineligible for CAR-T cell therapy can receive therapies as described for patients with non-high-risk relapse.

6 Managing Disease Recurrence After CAR-T Cell Therapy

Patients may experience disease relapse after CD19-targeted CAR-T cell therapy due to a low persistence of CAR-T cells, CD19 antigen loss, or downregulation of CD19 expression [100]. Potential mechanisms of resistance to CAR-T cell therapy also include CAR-T cell exhaustion, impaired expansion, and cytotoxicity [101]. To date, there is little information on how to best manage patients who experience disease relapse following CAR-T cell therapy and strategies for other salvage regimens. Salvage regimens (chemotherapy, radiation, and/or immunomodulatory drugs) are often used to manage disease recurrence following CAR-T cell therapy. The choice of salvage therapy depends on the prior treatment history and the timing and nature of the disease relapse. The role of hematopoietic stem cell transplant after CAR-T cell therapy in achieving long-term improvement in patients with r/r FL still needs to be explored. Retreatment of patients with CAR-T cell therapies is emerging as a promising option. In the ZUMA-5 study, patients retreated with axicabtagene ciloleucel as a fourth-line therapy demonstrated deep and durable responses (ORR 100%; CRR 77% for a median follow-up of 11.4 months; median DOR not reached; and 46% ongoing responses at data cutoff) with an acceptable safety profile [102]. In general, patients who experience disease relapse following CAR-T cell therapy should also consider enrolling in a clinical trial as there is no established standard of care in this setting; however, patients with persistent cytopenia following CAR-T cell therapy may not meet eligibility requirements for all clinical trials.

7 CAR-T Cell Therapy: Patient Identification, Challenges, and Clinical Needs

As treatment options for patients with r/r FL in the third-line setting or later continue to expand, integrating novel therapies in the treatment paradigm will require a better understanding of patient selection criteria. Currently, the criteria used to select patients who are likely to benefit from CAR-T cell therapy focus on patient-specific factors, including age, disease burden, previous therapies, comorbidities, organ function, and fragility. Patients with r/r FL are often of advanced age (i.e., >65 years) and are more likely to have comorbidities that could influence treatment choice. While results from the ELARA and ZUMA-5 clinical trials indicate that a high tumor burden may elevate risk of severe CRS [3, 40, 102, 103], therapies such as tisagenlecleucel can still be safely administered in these patients while following standard strategies for managing CRS. Patient selection can also be affected by logistical considerations, such as ease of access to the certified treatment center, the availability of caregiver support during key periods, and the ability to stabilize disease during CAR-T cell manufacturing. Ultimately, a long-term follow-up of the CAR-T cell therapy trials and future analyses of real-world outcomes will help to further define the characteristics of patients who benefit most from CAR-T cell therapies.

Additionally, with more treatment options in the third-line r/r FL setting, a unique clinical challenge of how to individualize the sequencing of therapy to achieve the best possible outcomes, as well as how to most effectively manage disease relapse in patients who have received CAR-T cell therapy, is arising. Following CAR-T relapse, treatment options include SCT, BsAbs, and salvage chemotherapy regimens. Currently, clinical trials supporting BsAb efficacy and safety have enrolled a subset of patients (10–40%) whose disease has relapsed following CAR-T cell therapy. The patients with prior CAR-T exposure achieved similar, albeit slightly lower, responses to patients with no prior CAR-T exposure, suggesting that BsAb therapy was safe and practical in patients with post-CAR-T disease relapse [104,105,106,107]. Currently, survival data with BsAbs are immature, and real-world evidence will be needed to assess response durability that will determine if BsAbs offer an alternative to CAR-T cell therapy in the future, given their preferable off-the-shelf access and safety profile. Long-term follow-up as well as real-world data are needed to help to identify the best treatment options for these patients. Planned/ongoing clinical studies will also examine the feasibility and benefit of re-treating patients with CAR-T cell therapy following a relapse [102].

8 Conclusions

Evolving treatment options for patients with r/r FL are highlighting the necessity for personalized treatment strategies that require careful consideration of previous treatment regimens and patient-related criteria, such as age, disease status, tumor burden, and comorbidities. Successful integration of CAR-T cell therapy into the treatment journey for patients with r/r FL will require identification of patients who are likely to benefit from CAR-T cell therapy. In addition, experience has demonstrated the importance of establishing a close collaboration between community oncologists and CAR-T treatment centers to achieve the best possible clinical outcomes for our patients. Proactively building this collaborative network also provides community physicians with a valuable resource for selecting the right treatment option for their patients and may help them better educate their patients on treatment options and the benefits of referral to authorized treatment centers. Future analyses of real-world data may help to inform the optimal sequencing of novel therapies in this patient population and provide insight on how best to address disease relapse following CAR-T cell therapy. Overall, as the benefits of CAR-T cell therapy in patients with r/r FL continue to accrue, we look forward to additional clinical trials that may further define their place in therapy.