Follicular lymphoma (FL), the second most common subtype of non-Hodgkin lymphoma (NHL) and the most common indolent NHL, accounts for 35 percent of NHLs in the U.S. FL arises from germinal center B cells, which are usually CD10 positive, CD5 negative, and CD20 and CD19 positive.
Approximately 85 percent of patients with FL have t(14;18), which results in overexpression of B cell leukemia/lymphoma 2 (BCL2), an oncogene that blocks apoptosis, leading to prolonged cell survival. Pathologic grading of FL (1, 2, 3A, 3B), which is based on an increasing number of centroblasts, is prognostic and predictive (Blood 2003;101:1149-1154). FL grade 3B is treated, as are aggressive lymphomas, with rituximab and anthracycline-based chemotherapy.
Although FL is not typically curable with conventional treatment, the prognosis for FL continues to improve, with current treatments extending median survival in excess of 14 years (Blood 2013;122:981-987). Tumor grade, tumor burden, and the FL International Prognostic Index (FLIPI-1, FLIPI-2) have been used to distinguish low-risk from high-risk disease. The model includes age, hemoglobin level, lymph node diameter, beta-2 microglobulin level, and the presence or absence of bone marrow involvement (J Clin Oncol 2009;27:4555-4562). The m7-FLIPI incorporates biological factors, such as mutations in seven genes (EZH2, ARID1A, MEF2B, EP300, FOXO1, CREBBP, and CARD11), identifying patients at high risk for poor outcomes to first-line therapy (Hematology 2017; doi:10.1182/asheducation-2017.1.358).
When to Initiate Treatment
Overall, 75-85 percent of patients with FL present with advanced disease (stage II bulky, III/IV) at diagnosis and have been historically managed with chemoimmunotherapy when treatment is indicated (Ann Oncol 2013;24:441-448).
When to start treatment rests largely on symptoms, the presence of significant cytopenias, and the threat of end organ damage. The presence of high tumor burden often aids in the decision to initiate therapy and has been proposed as an indicator by the Groupe d'Etude des Lymphomes Folliculaires (GELF) (J Clin Oncol 1997;15(3):1110-1117). GELF criteria include involvement of three or more nodal sites, each with diameter of 3 cm or more, any nodal or extranodal tumor mass of 7 cm or more, B symptoms, splenomegaly, pleural effusions or ascites, cytopenias (leukocytes <1.0 x 109/L and/or platelets <100 x 109/L), circulating lymphoma cells (>5.0 x 109/L malignant cells), and compressive symptoms. If a patient does not have significant symptoms and has low tumor burden, close observation is often recommended. This recommendation is based on prospective trials that have demonstrated no difference in overall survival when therapy is deferred.
Certainly, in clinical practice, a significant proportion of patients do not feel comfortable with the “watch and wait” approach, and initial single-agent therapy with rituximab is acceptable for patients with low tumor burden. A large international randomized trial comparing watchful waiting with initial treatment with rituximab suggested that initial treatment with rituximab may improve quality of life and allow postponement of cytotoxic chemotherapy (Lancet Oncol 2014;15(4):424-435).
The addition of the anti-CD20 monoclonal antibody rituximab to conventional chemotherapy has improved outcomes in FL, including response rate (RR), progression-free survival (PFS), event-free survival (EFS), and overall survival (OS).
The optimal standard chemotherapy regimen, or backbone, to partner with rituximab is unclear. R-CHOP (rituximab combined with cyclophosphamide, doxorubicin, vincristine, prednisone) was previously the most commonly used regimen in the U.S. prior to the introduction of the alkylating agent bendamustine. A phase III trial from the Study Group of Indolent Lymphomas (StiL) comparing six cycles of bendamustine plus rituximab (BR) to six cycles of R-CHOP demonstrated superior efficacy and reduced toxicity with BR (Lancet 2013;381(9873):1203-1210). In the patients with FL alone, the median PFS was significantly longer after BR than after R-CHOP (69.5 versus 31.2 months; HR 0.58), while OS did not differ significantly between groups at 10 years (70% vs. 66%).
A confirmatory phase III study (a follow-up to BRIGHT) also showed that BR was significantly superior to R-CHOP and R-CVP (rituximab combined with cyclophosphamide, vincristine, and prednisone) for PFS at 5 years (66% vs. 56%) for the group of indolent lymphomas as a whole. However, this difference lost statistical significance when the analysis was limited to FL (J Clin Oncol 2017;35(Suppl):abstract 7500). Once again, there was no difference in OS. The quality of life is somewhat better for patients taking BR because of the absence of alopecia, significant neuropathy, and steroid use. Limited data exists regarding the use of BR in patients with more clinically aggressive disease, such as histologic grade 3A or disease characterized by a high Ki-67 level (≥30%). In this subset of more aggressive FL, R-CHOP may be a preferred regimen.
For first-line treatment of patients with advanced FL, the FDA in November 2017 approved obinutuzumab in combination with chemotherapy, followed by obinutuzumab maintenance. The approval is based on data from the phase III GALLIUM study (N Engl J Med 2017;377:1331-1344), in which combining obinutuzumab with chemotherapy in first-line treatment, rather than relying on rituximab plus chemotherapy, reduced the risk of disease progression or death by 28 percent in patients with FL. There was no difference in OS and rates of histologic transformation at 3 years. The chemotherapy regimens used in GALLIUM were bendamustine (57%), CHOP (33%), or CVP (10%), and choice was based on the discretion of the physicians at each study location. One should note that fatal adverse events were higher overall in the obinutuzumab-treated patients (4.0%) than in the rituximab-treated patients (3.4%) and particularly higher when obinutuzumab was combined with bendamustine (5.6%). Rates were less than half that when either obinutuzumab or rituximab was combined with CHOP or CVP (≤2%).
The RELEVANCE trial is a phase III randomized study of lenalidomide plus rituximab (R2) versus chemotherapy plus rituximab, followed by rituximab maintenance, in patients with previously untreated FL of grades 1-3A. Findings recently presented at the 2018 ASCO Annual Meeting (J Clin Oncol 2018;36 (Suppl): abstract 7500) showed that neither coprimary endpoint of PFS or complete response/unconfirmed complete response (CR/CRu) favored R2. Efficacy was similar in both arms; however, safety data differed, with the chemotherapy plus rituximab arm registering higher levels of grade 3/4 neutropenia, febrile neutropenia, growth factor usage, nausea, vomiting, neuropathy, and alopecia than the R2 arm. Meanwhile, more frequent cutaneous reactions, tumor flare, and diarrhea were seen in R2. Primary investigator Nathan Fowler, MD, concluded that R2 is a novel immunomodulatory approach, and with its improved safety and tolerability over chemotherapy plus rituximab, R2 is a potential first-line chemotherapy-free option for patients with FL requiring treatment.
The question of whether to give maintenance rituximab after frontline rituximab plus chemotherapy was addressed in the phase III PRIMA trial. Patients with newly diagnosed FL who responded to initial treatment with rituximab plus chemotherapy were randomized to either observation or a single dose of rituximab every 2 months for 2 years.
At median follow-up of 36 months, the 2-year PFS in the maintenance rituximab arm was 75 percent, much higher than the 58 percent in the observation arm. This benefit was seen irrespective of the induction chemotherapy backbone. No difference in OS was observed. Maintenance rituximab was associated with a slightly higher incidence of grade 3/4 adverse events (24% vs. 17%) (Lancet 2011;377(9759):4-6).
Long-term follow-up of the PRIMA trial was presented at the 2017 American Society of Hematology Annual Meeting, and findings indicated that rituximab maintenance after induction immunochemotherapy provides a significant long-term PFS benefit over observation at 10 years (51% vs. 35%) (Blood 2017;130(Suppl1):abstract 486).
Therapy for Relapsed & Refractory Disease
General Therapeutic Considerations
Relapse of FL commonly presents as asymptomatic enlargement of the lymph nodes, liver, or spleen. Asymptomatic patients can also be closely observed, and treatment should be based on the modified GELF criteria as in first-line therapy.
If histologic transformation or a change in biology is suspected (rising lactate dehydrogenase levels, disproportionate growth of single site, new extranodal disease, new B symptoms), a low threshold for a new biopsy is recommended, and PET scanning using fludeoxyglucose F 18 (FDG) may help identify suspicious sites of transformation if intense FDG avidity is seen.
Patients with relapsed or refractory FL will find physicians have no standard therapy to offer, and patients should be encouraged to participate in clinical trials. The choice of therapy at relapse should take into consideration the patient's prior treatment and duration of response to initial therapy and his or her age, comorbidity, and goals for therapy.
Patients with early relapse within 2 years of initial diagnosis have poor outcomes. For example, the 5-year survival for patients with early relapse after rituximab plus CHOP was only 50 percent, whereas the rate was greater than 90 percent for patients who did not experience early relapse (J Clin Oncol 2015;33(23):2516-2522). These patients should be considered for more aggressive treatment, such as undergoing autologous hematopoietic stem cell transplant, or encouraged to participate in clinical trials.
For patients with later relapse, chemoimmunotherapy with an alternate chemotherapeutic or biologic backbone followed by maintenance is reasonable and can continue to provide long-term disease control. It should be noted that radioimmunotherapy has demonstrated response rates of 60-80 percent; however, it is not recommended for patients with poor bone marrow reserve or high marrow tumor burden (Cancer 2007;109(9):1804-1810). This modality is also not commonly used because of the complexity of administration.
Targeting the Microenvironment
Over the last decade, biological advances have led to the discovery of a plethora of potential therapeutic targets and the development of many novel agents that target tumor cell surface molecules, intracellular signaling pathways, reactive cells of the microenvironment, and immune checkpoints.
The tumor microenvironment is increasingly recognized as crucial to sustaining malignant B-cell survival and growth, subclonal evolution, and drug resistance (Hematology 2017;2017:610-617). FL has long been considered as immune responsive based on high responses to anti-CD20 antibodies and trials involving immunomodulatory drugs. Novel agents, whether approved or in development, aim to target the lymphoma microenvironment to stimulate meaningful responses.
- Immunomodulatory drugs. Lenalidomide is an immunomodulatory agent activating natural killer and T cells, and its administration results in apoptosis of neoplastic B cells and activates complementary mechanisms with rituximab (J Clin Oncol 2015;33(25):2803-2811). Both agents have direct effects on lymphoma cells, as well as immunostimulatory effects that recruit the host immune system to attack the lymphoma. The MAGNIFY study was a phase IIIB randomized study of R2 followed by maintenance in relapsed/refractory NHL. The ORR for all FL was 66 percent, with 38 percent CR/CRu. The ORR in the “double-refractory” and early relapse groups was 45 percent and 47 percent, respectively, including a CR/CRu of 21 percent for both groups. Double-refractory refers to disease that within 6 months is unresponsive to or fails to continue responding to therapy with rituximab and an alkylating agent.
- PI3K inhibitors. The first PI3K-δ inhibitor approved for the management of FL was idelalisib, after a phase II study demonstrated an objective response rate (ORR) of 56 percent, a CR of 14 percent, and a median duration of response of 10.8 months in a heavily pretreated population. This population included 100 percent of patients who were identified as having double-refractory disease (N Engl J Med 2014;370:1008-1018). However, fatal and/or serious hepatotoxicity occurred in 16-18 percent of patients treated with idelalisib, and concerns of toxicity in combination have hindered uptake of this agent in the FL community. There is however promise for the next generation of PI3K inhibitors.
Copanlisib is an IV pan-PI3K inhibitor approved in September 2017 for relapsed FL and has more selective activity against alpha and delta subunits than do other inhibitors. A phase II clinical trial (CHRONOS-1) demonstrated a 59 percent ORR and a 14 percent CR and was well-tolerated in general (J Clin Oncol 2017;35(Suppl):abstract 7535). Grade 3 and 4 hepatotoxicity was low (1%). Since the alpha isoform is broadly expressed and involved in insulin signaling and angiogenesis, hyperglycemia (all grades, 54%) and hypertension (all grades, 35%) were increased with copanlisib. Phase III trials of copanlisib in combination with rituximab (CHRONOS-3) and in combination with chemoimmunotherapy (CHRONOS-4) are ongoing in patients who have relapsed.
Duvelisib is an oral PI3K-δ and PI3K-γ inhibitor, which was granted priority review in April 2018 for relapsed/refractory chronic lymphocytic leukemia (CLL), small lymphocytic lymphoma (SLL), and FL. In the phase II DYNAMO study, duvelisib demonstrated an ORR of 41 percent in patients with FL, including patients with heavily pretreated and double-refractory disease (Hematol Oncol 2017;35(Suppl 2):69-70).
ME-401 is a novel and potent oral PI3K-δ that is highly selective for the delta isoform. Initial results of a phase I dose-escalation study of ME-401 in relapsed or refractory FL, CLL, and SLL were presented at the 2018 ASCO Annual Meeting. The ORR in FL was impressive at 86 percent (n=30), including a 100 percent RR in early progressors. At the median follow-up of 8 months, no dose-limiting toxicities were observed thus far at the first three dose levels. Grade 3 diarrhea was seen in 19 percent; however, hepatotoxicity was not common. The half-life of 28 hours allows for once-daily dosing and a unique intermittent dosing schedule of days 1-7 every 28 days is also currently being explored. (An abstract online presents background and data collected at a median follow-up of 20 weeks [J Clin Oncol 2018;36[Suppl]:abstract 7519].)
- Bruton tyrosine kinase inhibitors. Ibrutinib as a single agent at 560 mg daily in relapsed/refractory FL recently demonstrated an ORR of 21 percent in an open-label, single-arm, phase II study (DAWN). Although the study did not achieve its primary objective, secondary endpoints such as median duration of response of 19.4 months and lymphoma symptom resolution rate of 67 percent suggest benefits of this therapy in some patients (J Clin Oncology 2018; doi:10.1200/JCO.2017.76.8853). A combination of ibrutinib and rituximab in untreated FL have shown an ORR of 85 percent and a CR of 35 percent (Blood 2015;126(23):abstract 470). At the 2018 ASCO Annual Meeting, acalabrutinib and rituximab demonstrated a 92 percent ORR in untreated FL, and a 39 percent ORR in relapsed/refractory FL. (See background and preliminary data at J Clin Oncol 2018;36(Suppl):abstract 7549.)
- BCL2 inhibitors. Venetoclax as a single agent in relapsed FL demonstrated a modest ORR of 38 percent, which was lower than expected given the role of BCL2 in FL lymphomagenesis (J Clin Oncol 2017;35(8):826-833). Additional investigations including combination therapy to augment response rates and durability are ongoing.
- Immuno-oncology targets. Anti-CD47 antibody Hu5F9-G4 is a first-in-class macrophage immune checkpoint inhibitor targeting the dominant macrophage checkpoint CD-47, which is a “do not eat me” signal on cancers that enables macrophage immune evasion. Cancer cells express prophagocytic (“eat me”) signals, while most normal cells do not. A 5F9/CD47 blockade induces anti-tumor activity in over 25 tumor models, including NHL. The extrinsic “eat me” signals provided by rituximab through the Fc receptor enhance 5F9 activity. CD47 blockade takes the foot off the brakes, while rituximab puts the foot on the accelerator, leading to maximal tumor phagocytosis. The initial phase Ib dose escalation results of 5F9 in combination with rituximab in relapsed/refractory B-cell NHL was reported at the 2018 ASCO Annual Meeting (J Clin Oncol 2018;36(Suppl):abstract 7504). In seven patients with FL refractory to prior rituximab, the ORR was 71 percent (5/7) with a CR of 43 percent (3/7). Median duration of response has not been reached, with at least one patient in CR for more than 14 months. The 5F9 was well-tolerated in combination with rituximab with no maximum tolerated dose achieved. On-target anemia was transient and significantly mitigated by priming and maintenance dosing. No patient developed autoimmune adverse events. Recently, the FDA granted Hu5F9-G4 fast-track designation for relapsed diffuse large B-cell lymphoma (DLBCL) and FL.
- Bispecific T-cell–engaging antibody. Blinatumomab is a bispecific T-cell–engaging antibody that transiently links CD3-positive T cells to CD19-positive B cells, inducing T-cell activation and proliferation, and tumor cell lysis. Blinatumomab was evaluated in a phase I trial in relapsed/refractory NHL, including 28 patients with FL. The ORR in the heavily treated population with FL was encouraging at 80 percent (J Clin Oncol 2016;34(10):1104-1111). Neurologic adverse events likely from cytokine release were the most common dose-limiting toxicities.
- Chimeric antigen receptor T-cell therapy. CAR T-cell therapy is an investigational therapy in FL that genetically modifies the patient's own T lymphocytes to target lymphoma cells. Two agents, axicabtagene ciloleucel and tisagenlecleucel, are now FDA-approved for relapsed refractory DLBCL. Data regarding the use of CAR T-cell therapy in FL comes from small prospective trials. In a single-center trial of CTL019 in patients with double-refractory FL, 10 of 14 patients attained a CR by 3 months (N Engl J Med 2017;377(26):2545). Among patients who achieved CR, all but one remained in CR at a median follow-up of 2 years. Severe cytokine release syndrome occurred in five patients (18%). Serious encephalopathy occurred in three patients (11%): two cases were self-limiting and one case was fatal.
Unmet Needs & Future Direction
In the era of precision and personalized medicine, FL lags behind some of its hematologic counterparts, which benefit from biomarkers that guide therapy. In CLL, for instance, fluorescence in situ hybridization or immunoglobulin heavy-chain variable mutational testing results can help tailor selection of frontline therapy. Testing for 17p deletion will identify patients who will derive more benefit from therapy with the Bruton tyrosine kinase inhibitor ibrutinib than with conventional chemotherapy. Unfortunately, routinely available prognostic and predictive biomarkers are not yet available in the management in FL.
The most powerful predictors of outcome today are the quality of response to initial therapy, as assessed by end-of-treatment PET imaging, or by durability of the first remission. Ongoing research will someday identify validated prognostic biomarkers capable of identifying high-risk patients and reliable predictive markers to select subgroups most likely to benefit from novel approaches. Such tools will prevent early relapse and further improve survival in FL.
Developing combinations that target driver mechanisms within lymphoma cells, minimizing drug resistance along pathways, and harnessing the potential of the immune system will be paramount to improving outcomes in FL.
ALAN TAN, MD, is Director of Hematologic Malignancies and Clinical Research at Cancer Treatment Centers of America, Phoenix.
Read More Articles on Leukemia & Lymphoma
Oncology Times offers a Lymphoma & Leukemia Resource Center with articles all in one place. Visit http://know.lww.com/lymphoma-leukemia.html to read the latest research.