Title: Long‐term efficacy and safety of chemotherapy‐free first‐line iodine‐131‐rituximab radioimmunotherapy of follicular lymphoma
Abstract: Advances in the treatment of follicular lymphoma (FL) have fostered the hope that some patients may be cured with the use of immunochemotherapy. In patients who are destined for long-term survival and possible cure, attention is turning toward reducing toxicity and improving quality of life.1, 2 The relative differences in morbidity and mortality attributable to induction regimen-related adverse events (AEs), now assume much greater significance.2-6 Current up-front chemoimmunotherapy combinations are accompanied by AEs in up to 97% and serious AEs in up to 50% of FL patients treated first-line.3-7 The single-centre phase II INITIAL (ANZCTR 12607000153415) study using iodine-131-rituximab (131I-RIT) delivered in an outpatient setting without chemotherapy, demonstrated a complete metabolic remission (CR) in 82% of patients at three months8 and a median time-to-next-treatment (TTNT) not reached (median follow-up of 4·25 years).8 Compared with chemoimmunotherapy combinations, a single episode of treatment with 131I-RIT resulted in no significant serious or fatal adverse event during induction/maintenance. The relative absence of toxicity of 131I-RIT supported the previously demonstrated safety and efficacy of single-agent first-line RIT for FL.9 However, due to limited availability, a lack of randomised studies and concern regarding the long-term haematologic toxicity and oncogenic potential of radionuclide exposure, RIT remains largely investigational.10 We now provide updated data on the efficacy and safety of 131I-RIT induction therapy of FL collected over a decade of follow-up of patients treated on the phase II INITIAL study.8 The study design has previously been described in detail.8 Briefly, patients with untreated, advanced, symptomatic grade 1 or 2 FL were eligible. Staging was determined by baseline 18F-fluorodeoxyglucose positron emission tomography/computed tomography (18F-FDG PET/CT) and bone marrow biopsy. Treatment with a standard course of immunotherapy comprising four doses of 375 mg/m2 rituximab at weekly intervals was administered to each patient. The second and third administration of rituximab immediately preceded the tracer and therapy doses of 131I-rituximab, respectively. Administered therapeutic activity of 131I-rituximab was determined by individual prospective dosimetry using quantitative whole-body gamma imaging of a tracer activity 250 megabecquerel (MBq) 131I-rituximab. The prescribed radiation absorbed dose to the whole body was fixed at 0·75 Gray (Gy) in each patient. Treatment response was determined by 18F-FDG-PET/CT scans performed three months after 131I-RIT induction and reported according to both the revised International Harmonization Project (IHP) 11 and Deauville response criteria.12 Patients who achieved a treatment response received standard maintenance treatment comprising four 375 mg/m2 doses rituximab, three-monthly for one year. Follow-up and use of imaging studies thereafter was at the discretion of the referring physician. Overall survival (OS) and TTNT were end-points of interest for the current analysis and were estimated according to the Kaplan–Meier method. A Cox proportional hazards model was used to assess the prognostic significance of FLIPI and disease response at three months with respect to OS and TTNT. Cumulative incidence of death from second malignancies was calculated using the non-parametric Nelson–Aalen estimator. Persistent haematologic toxicity (PHT) and/or immunoparesis were defined as sustained cytopenias or immunoglobulin G (IgG) < lower limit of normal (LLN), present on two or more studies performed at least six months apart, graded as per the common terminology criteria. Recurrent infection was defined as ≥3 infections per year requiring systemic antibiotics and/or meeting criteria for immunoglobulin replacement [intravenous immunoglobulin (IVIG)].13 Sixty-eight patients were enrolled between April 2006 and December 2013 (Table I). Cut-off date for the current analysis was 7 July 2021. All patients completed the induction phase of therapy. One patient experienced early disease progression and commenced salvage immediately prior to the formal three-month response assessment. The overall completion rate for the study was therefore 98·5%. Two patients were lost to long-term follow-up due to relocation overseas. At a median follow-up of 9·5 years (95% CI: 0·85–14·6), 18 (26·5%) patients have relapsed; 2 (11%) were with high-grade transformation and 15 (83·3%) have required salvage therapy. The median OS and TTNT have not been reached. The estimated 10-year OS and TTNT was 80% (68·27–87·98) and 76·7% (64·31–85·29) respectively (Fig 1). The original study reported a three-month overall response rate (ORR) in 67 (98·5%) patients; 56 (82·3%) CR, 11 (16·2%) partial remission (PR) and 1 (1·5%) progressive disease (PD; clinically evident). Complete remission defined by three-month PET/CT response assessment was the only significant predictor of an improved estimated 10-year OS [hazard ratio (HR), 0·016; 95% CI, 0·061–1·29 P = 0·02] and TTNT (HR, 0·33; 95% CI, 0·076–1·41, P = 0·03). Current analysis did not observe any clinically significant PHT; five (8%) patients developed grade one lymphopenia and one (1·6%) immunoparesis. Three patients developed a second malignancy (one brain tumour, one metastatic breast cancer, one metastatic melanoma) at 3·2, 3·5 and 7 years respectively, with an estimated 10-year cumulative incidence of 4·92% (95% CI, 1·58–15·30). A total of 12 (17·6%) deaths have been recorded; three (25%) disease-related, three (25%) from their secondary malignancy, two (16·7%) from an unrelated infection and four (33·3%) from other medical causes. No instance of myelodysplasia/acute leukaemia (MDS/AL) has been observed. This is the first report of long-term patient outcomes following first-line 131I-RIT of FL. The results demonstrate sustained treatment-free remissions with minimal long-term toxicity and excellent survival outcomes, comparing favourably to first-line yttrium-90-ibritumomab RIT of FL.9, 14 The current therapeutic landscape for FL is dominated by chemoimmunotherapy induction followed by antibody maintenance for up to two years. Despite superior progression-free survival (PFS) observed with obinutuzumab–bendamustine, an OS advantage over alternative combinations has not been demonstrated and extended follow-up is pending.15 The 10-year OS estimate of 80% in our study is similar to that reported in the long-term PRIMA update (median follow-up nine years).16 Despite a similar proportions of high-risk patients, the 10-year TTNT estimate of 76% in our analysis versus 53% for the PRIMA maintenance arm16 is likely attributed to our small samples size and single-arm design. Nevertheless, our results support the long-term efficacy of 131I-RIT induction. Given the generally favourable prognosis and rising age of incidence of FL there is concern for treatment-related morbidity/mortality.1, 7, 17 In a 10-year follow-up real-world study of 1 654 newly-diagnosed FL patients treated in the rituximab era treatment-related death accounted for 17% of fatal outcomes, with deaths resulting from infection (8%), MDS/AL (5%) and cardiotoxicity (2·4%).1 In the GALLIUM Study the bendamustine cohort experienced an increased risk of grade 3–5 infections and higher incidence of second malignancies, including MDS/AL.3, 7 Similar risks are highlighted in updated follow-ups from the BRIGHT, StiL and FOLL05 studies.4, 5, 17 First line single modality 131I-RIT also appears to ameliorate the increased risks associated with combination/consolidation RIT strategies.18-20 In our long-term follow-up of 131I-RIT we have recorded no case of treatment-related mortality. The estimated 10-year cumulative incidence of fatal second malignancy after 131I-RIT was 5%, compared with up to 15% reported with chemoimmunotherapy.4, 5, 7, 17 Chemoimmunotherapy combinations also vary significantly with respect to administration protocols/treatment duration. In this context, the 131I-RIT protocol carries a minimal day patient burden. The absence of an immune-chemotherapy arm in our study does not allow objective direct comparative evaluation. It does, however, address the assignment of lesser weight to outcome measures derived from surrogates for survival such as PFS by incorporating other relevant clinical measures including TTNT, long-term survival and toxicity. Iodine-131-rituximab radioimmunotherapy is a safe, chemotherapy-free treatment, with sustained efficacy and minimal toxicity burden. The exact role of 131I-RIT within the FL therapeutic landscape remains to be defined. However, where available it may be considered a preferred alternative first-line treatment, especially in patients at increased risk of chemotherapy-related toxicity. MK, JHT and ADM conceptualised and designed the paper. MK, GZ and JTM performed the data collection. MK performed the primary data analysis and wrote the first draft of the manuscript. WBM helped conceive the study, assisted with central review of radiology and data. JHT and ADM reviewed and co-wrote the manuscript. All authors reviewed and approved the final version of the manuscript. No grant funding, or pharmaceutical industry support, was solicited or received. None of the authors has any potential conflict-of-interest relationship to disclose.