Long-term safety for patients with tisagenlecleucel-treated relapsed/refractory diffuse large B-cell lymphoma

TO THE EDITOR:

Chimeric antigen receptor (CAR) T-cell therapy has promising outcomes in patients with relapsed/refractory (R/R) diffuse large B-cell lymphoma (DLBCL). The incidence, grading, and management of acute adverse events (AEs) have been described for CAR T-cell therapies.^1-5  Reports of long-term (LT) AEs being treated with CAR T-cell therapy are emerging,^2,6-8  and new management practices are being established.^9,10  In the pivotal, global, single-arm, phase 2 JULIET trial, tisagenlecleucel demonstrated efficacy and manageable safety in adult patients with R/R DLBCL⁵  and ongoing durable efficacy at a median of 40.3 months of follow-up.¹¹  Here, we report the LT safety profile of tisagenlecleucel from the JULIET trial.

Eligibility and end points for the JULIET trial (NCT02445248) were described previously.⁵  Patients (age 18 years or older) with aggressive B-cell lymphomas who received ≥2 previous lines of therapy were eligible.⁵  All AEs were summarized by using the maximum grade recorded. Responders were defined as patients with best overall response of either complete or partial response. Efficacy and short-term AEs have also been described previously.⁵  The type, frequency, and severity of LT AEs (ie, those that occurred or persisted beyond 90 days or occurred beyond 2 years after infusion) were recorded using Common Terminology Criteria for Adverse Events v4.03. Cytopenia grade was determined by measuring lymphocytes, neutrophils, hemoglobin, and platelets. LT cytopenias occurred or persisted beyond 90 days.

Hypogammaglobulinemia was defined as immunoglobulin G (IgG) <4 g/L and B-cell aplasia was defined as <0.2 CD19⁺ cells per microliter. Data regarding CD4 count are not available for this patient population but can be investigated in future studies. Administration of intravenous immunoglobulin (IVIg) was at the physician’s discretion. IgG, IgM, and IgA levels were evaluated at baseline, on days 14 and 28, at months 3, 6, 9, and 12, and at the end of follow-up (5 years), together with the use of IVIg and clinical outcomes. Baseline measurements were defined as the last measurements before infusion. Infections after 12 months were recorded if they required anti-infective treatment or led to significant disability, hospitalization, and/or surgery. Categorical data (eg, sex) were summarized as frequency counts and percentages, and continuous data (eg, age) were summarized by descriptive statistics (eg, mean).

Resolution of grade 3 to 4 cytopenias to grade ≤2 and onset of remission to B-cell recovery (≥1% B cells in white blood cells or ≥3% B cells in lymphocytes) was analyzed via the Kaplan-Meier method and was reported from infusion until disease progression, at initiation of new anticancer treatment, or at last available follow-up or death, whichever came first. All 95% confidence intervals (CIs) for Kaplan-Meier estimates were calculated using log-log transformation within PROC LIFETEST (SAS v9.3).

The JULIET trial was designed and sponsored by Novartis Pharmaceuticals, was approved by the institutional review board at each participating institution, and was conducted according to the Declaration of Helsinki. Data were analyzed and interpreted by the sponsor and the authors. The authors ensured adherence of the study to the protocol, which is available in the supplemental Data. The safety analysis included 115 patients who received infusions of tisagenlecleucel. As of 1 July 2019, median follow-up was 32.6 months (maximum, 44.9 months), and 60 patients (52%) had a response. Baseline characteristics of patients have previously been described.⁵ 

Of the responders, 16 (27%) of 60 had LT cytopenia (ie, grade ≥3), including 2 (3%) with anemia, 8 (13%) with thrombocytopenia, 7 (12%) with lymphopenia, and 9 (15%) with neutropenia (Table 1). Among responders with LT cytopenias, 14 (88%) of 16 received 2 to 4 (range, 1-6) previous lines of therapy, similar to the 31 responders without LT cytopenias. Eleven responders (69%) with LT cytopenias received red blood cell and/or platelet transfusions after treatment with tisagenlecleucel, of which 4 (25%) received transfusions and granulocyte colony-stimulating factor after 90 days.

For responders, median time to resolution of cytopenia to grade ≤2 was 3.5 months (95% CI, 3.19 months to not estimable [NE]) for anemia, 6 months (95% CI, 3.1-11.9 months) for thrombocytopenia, 4 months (95% CI, 3.0 months to NE) for lymphopenia, and 4 months (95% CI, 3.1-10.2 months) for neutropenia. Most cytopenias resolved by month 12 after infusion, except for 21% of patients who had LT lymphopenia (Table 1). Three nonresponders from a total of 55 patients had LT cytopenias, including 1 (2%) with anemia, 1 (2%) with thrombocytopenia, and 1 (2%) with lymphopenia; no resolution was observed. In all, 74% of patients (85 of 115) and 73% of responders (44 of 60) had B-cell aplasia before infusion (supplemental Table 1). For responders, median time to B-cell recovery was 11 months (95% CI, 5.3-16.9 months; supplemental Figure 1), and 30 responders had B-cell recovery.

Hypogammaglobulinemia occurred in 62 patients (54%) (35 responders, 27 nonresponders), 29 had baseline IgG levels ≥4 g/L, and 27 (43.5%) received IVIg. Sixteen responders (46%) did not have hypogammaglobulinemia before infusion. Among responders, the median duration of hypogammaglobulinemia was 742 days (range, 0-1132 days); 20 patients (57%) received IVIg after infusion. Doses of IVIg were heterogeneous and followed local guidelines (supplemental Table 2). Infections occurred in 21% of patients (24 of 115), and 30% of patients (34 of 115) did not have hypogammaglobulinemia.

Grade 3 to 4 infections occurred in 17% of responders (10 of 60) ≤8 weeks, in 21% of responders (12 of 58) between 8 and 52 weeks, and in 14% of responders (6 of 42) >1 year after infusion. In 60 patients with grade 3 to 4 infections after infusion, 6 (10%) had pneumonia, 3 (5%) had urinary tract infections, and 3 (5%) had general infections. Infections were predominantly bacterial and viral; 1 patient had cytomegalovirus and 2 patients had herpes simplex virus.

Among responders with LT cytopenias, grade 3 infections occurred in 1 patient (7%) at ≤8 weeks, in 3 patients (20%) between 8 and 52 weeks, and in 1 patient (8%) >1 year after infusion. Among responders without LT cytopenias, grade 3 infections occurred in 5 patients (18%) at ≤8 weeks, and grade 3 to 4 infections occurred in 8 patients (29%) between 8 and 52 weeks and in 5 patients (20%) >1 year after infusion (Table 2). Grade 3 infections occurred among 29 patients with LT neutropenia: for responders, 1 patient (17%) at ≤8 weeks, 0 patients between 8 and 52 weeks, and 0 patients >1 year after infusion; for nonresponders, 6 patients (26%) at ≤8 weeks, 2 patients (12.5%) between 8 and 52 weeks, and 0 patients >1 year after infusion.

Of patients who received an infusion, 8% percent (9 of 115) reported secondary malignancies (SMs), including 1 patient with myelodysplastic syndrome (MDS) and 1 with acute myeloid leukemia (AML). Among SMs, 78% were grade ≥3 (non–life-threatening to acute life-threatening) and ∼50% occurred >1 year (range, 1.1-1.6 years) after infusion. The estimate for cumulative incidence at 42 months was 7.7% (95% CI, 3.6%-13.9%) for SMs and 58.6% (95% CI, 48.4%-67.4%) for death as a result of any cause. Considering only the patients who achieved a complete or partial response, 13% (8 of 60) developed SMs. This apparently higher incidence in responding patients can be explained by the prolonged survival and the cumulative toxicity of previous therapies in this subgroup compared with nonresponders. All patients with an SM had received at least 1 and up to 5 lines of previous therapies, including 4 patients who received autologous stem cell transplantation (ASCT) before infusion with tisagenlecleucel (supplemental Table 3).

Patients with R/R DLBCL treated with tisagenlecleucel had manageable LT AEs. Sixteen responders (14%) of a total of 115 patients had LT cytopenias lasting ≥90 days. Although mechanisms underlying LT cytopenias are unknown, most LT cytopenias resolved by 12 months. Patients with DLBCL who receive rituximab¹²  are at risk for de novo hypogammaglobulinemia and exacerbated baseline hypogammaglobulinemia.¹³  In the JULIET trial, 98% of the patients received previous treatment with rituximab,¹⁴  and 53% had hypogammaglobinemia before they received an infusion of tisagenlecleucel. Consistent with other studies,^2,6,7  we observed low rates of severe or opportunistic late infections; few responders had LT infections. Given the variability in management and the small number of patients, it is difficult to draw conclusions regarding any correlation between infections and cytopenias or hypogammaglobulinemia and use of IVIg. However, the low infection rate suggests that treatment centers are appropriately managing hypogammaglobulinemia, cytopenias, and infections.

Data from the California Cancer Registry showed that patients diagnosed with DLBCL between 2001 and 2012 had a 5-year cumulative incidence of ∼5% for SMs.¹⁵  The 15-year cumulative incidence of patients developing SMs after ASCT was ∼10%.^16,17  Here, SMs occurred in 8 patients (7%) who had received an infusion, similar to a previous tisagenlecleucel pilot trial with 5 years of follow-up.⁸  MDS and AML could be related to the number and type of previous therapies (eg, ASCT).⁶  Over a 3.74-year follow-up period in the JULIET trial, prostate cancer was diagnosed in 3 of 115 males (0.8% annual incidence) and breast cancer was diagnosed in 1 of 115 females (0.2% annual incidence). These incidence rates are lower than the annual incidence of prostate cancer (7.3%) and breast cancer (11.7%) in an age-matched population.¹⁸  Risk of SMs and LT AEs after CAR T-cell therapy needs further elucidation with longer follow-up from clinical trials like the JULIET trial.

Acknowledgments: The authors thank Jacqueline R. Ward of Health Care Consultancy Group who provided assistance with medical writing.

This work was supported by Novartis Pharmaceuticals.

Contribution: G.S., S.J.S., and R.T.M contributed to the study design; U.J., C.S.T., P.B., J.P.M., M.J., E.K.W., S.J., C.A., S.R.F., J.R.W., I.F., P.J.H., S. Mielke, T.T., G.S., S.J.S., F.H., R.T.M., S. Mayer, S. Makita, M.J.K., M.G., N.W.-J., K.K., P.C., H.G., and M.R.B. enrolled and treated patients and gathered data; U.J., G.S., S.J.S., R.T.M., S.C., A.A., C.L.-B., and M.R.B, accessed and verified the study data; and all authors analyzed and interpreted the data, participated in writing and reviewing this article, and approved the final submitted version.

Conflict-of-interest disclosure: U.J. reports grants and personal fees from Novartis Pharmaceuticals, Gilead Sciences, Celgene/Bristol Myers Squibb, Janssen, Roche, and Miltenyi Biotec and funding from the Innovative Medicines Initiative 2 Undertaking (grant 945393) (this Joint Undertaking received support from the European Union’s Horizon 2020 research and innovation program and the European Federation of Pharmaceutical Industries and Associations). C.S.T. reports honoraria from AbbVie, BeiGene, Janssen, Novartis Pharmaceuticals, and Roche and research support from AbbVie and Janssen. E.K.W. reports consulting fees, honoraria, travel expenses, membership on a board of directors or advisory committee, and research funding from Novartis Pharmaceuticals; membership on a board of directors or advisory committee and research funding from Juno Therapeutics; is a member of the National Cancer Institute Leukemia Steering Committee; and is an advisor or consultant for and reports equity ownership in Cambium Medical Technologies and Cambium Oncology. S.J. reports research support from Kite Pharma and Novartis Pharmaceuticals and personal fees from Juno Therapeutics, Kite Pharma, and CRISPR Therapeutics. C.A. is an advisor or consultant for Gilead Sciences, Kite Pharma, Karyopharm Therapeutics, Atara Biotherapeutics, Incyte, TG Therapeutics, and Epizyme; reports research support from Novartis Pharmaceuticals, GlaxoSmithKline, Amgen, Juno Therapeutics, Celgene, and Merck; and reports travel support from Gilead Sciences and Kite Pharma. S.R.F. reports personal fees from Novartis Pharmaceuticals. J.R.W. is an advisor or consultant for Celgene, Curis, Genentech, Janssen, Juno Therapeutics, Kite Pharma, MorphoSys, and Novartis Pharmaceuticals and reports research support from Celgene, Curis, Forty Seven, Genentech, Janssen, Kite Pharma, Novartis Pharmaceuticals, and Unum. I.F. is an advisor or consultant for AbbVie, AstraZeneca, Bristol Myers Squibb, BeiGene, Gilead Sciences, Janssen, Novartis Pharmaceuticals, Roche, and Seattle Genetics and reports honoraria from AbbVie, Gilead Sciences, Janssen, Novartis Pharmaceuticals, Roche, and Seattle Genetics. P.J.H. reports travel support from Celgene, Janssen, La Jolla Pharmaceutical, and Novartis Pharmaceuticals. S. Mielke is an advisor or consultant for Novartis Pharmaceuticals; reports speaker fees from Celgene, Kiadis Pharma, and Miltenyi Biotec; reports participation in an expert panel for Bellicum Pharmaceuticals and Kite Pharma/Gilead; reports participation in data and safety monitoring boards for Miltenyi Biotec and DCPrime; reports leadership or fiduciary role for Sweden’s CAR T Cell Competence Network (SWECARNET); and reports travel support from Celgene, Kiadis Pharma, Kite Pharma/Gilead, and Miltenyi Biotec. T.T. is an advisor or consultant for Merck, Novartis Pharmaceuticals, and Takeda Pharmaceuticals; has received honoraria from Bristol Myers Squibb, Fuji Pharma, Kyowa Kirin, Merck, Nippon Shinyaku, Pfizer, Takeda Pharmaceuticals, and Teijin Pharma; reports research support from Astellas, Chugai Pharma, Kyowa Kirin, Novartis Pharmaceuticals, and Sanofi; reports grants from the Japan Society for the Promotion of Science and Japan Science and Technology Agency; and has received assistance with manuscript preparation for Janssen and Novartis Pharmaceuticals. G.S. is an advisor or consultant for AbbVie, Allogene Therapeutics, Autolus Therapeutics, BeiGene, Celgene, Epizyme, Genmab, Janssen, Kite Pharma/Gilead, Merck, MorphoSys, Novartis Pharmaceuticals, Roche, and VelosBio and reports honoraria from AbbVie, Amgen, Bristol Myers Squibb, Celgene, Epizyme, Janssen, Kite Pharma/Gilead, Merck, MorphoSys, Novartis Pharmaceuticals, and Roche. S.J.S. is an advisor or consultant for Acerta Pharma, Allogene Therapeutics, AstraZeneca, BeiGene, Celgene/Juno Therapeutics, Genentech/Roche, Loxo Oncology, Novartis Pharmaceuticals, and Tessa Therapeutics; has received honoraria from Acerta, Allogene Therapeutics, AstraZeneca, BeiGene, Celgene, Genentech/Roche, Loxo Oncology, Novartis Pharmaceuticals, Nordic Nanovector, Pfizer, and Tessa Therapeutics; has participated in steering committees for AbbVie, Celgene, Novartis Pharmaceuticals, Juno Therapeutics, Nordic Nanovector, and Pfizer; reports research support from AbbVie, Acerta Pharma, Celgene/Juno Therapeutics, Zhejiang DTRM Biopharma, Genentech, Incyte, Merck, Novartis Pharmaceuticals, Portola Pharmaceuticals, and TG Therapeutics; and holds a patent with Novartis Pharmaceuticals. F.H. is a consultant for AbbVie and Magenta Therapeutics. R.T.M. is an advisor or consultant for AlloVir, Artiva Biotherapeutics, CRISPR Therapeutics, CytoDyn, Incyte, and Novartis Pharmaceuticals; reports honoraria from Bristol Myers Squibb/Celgene, Incyte, Intellia Therapeutics, Kite Pharma, Omeros, Orca Biosystems, and PACT Pharma; research support from Juno Therapeutics and Novartis Pharmaceuticals; participation in data and safety monitoring boards for Athersys and Novartis Pharmaceuticals; and holds a patent with Athersys. S. Mayer is an advisor or consultant for Omeros. S. Makita is a consultant for Celgene/Bristol Myers Squibb, Daiichi-Sankyo, and Takeda Pharmaceuticals and reports honoraria from Celgene/Bristol Myers Squibb, CSL Behring, Eisai, Chugai Pharmaceutical, Novartis Pharmaceuticals, SymBio Pharmaceuticals, and Takeda Pharmaceuticals. M.J.K. is an advisor or consultant for and has received honoraria and travel support from Amgen, Bristol Myers Squibb, Celgene, Janssen, Kite Pharma/Gilead, Merck, Miltenyi Biotec, Novartis Pharmaceuticals, and Roche and reports research support from Celgene, Roche, and Takeda Pharmaceuticals. N.W.-J. is an advisor or consultant for ADC Therapeutics, Bayer, Epizyme, Gilead Sciences, Grünenthal, Janssen, Karyopharm Therapeutics, and Regeneron; and reports research support from Roche/Genentech, ADC Therapeutics, Astex Therapeutics, Juno Therapeutics, and Regeneron Pharmaceuticals. K.K. is an advisor or consultant for AbbVie, AstraZeneca, Celgene, Chugai Pharmaceutical, Eisai, Janssen, and Novartis Pharmaceuticals; has received honoraria from Celgene, Chugai Pharmaceutical, Janssen, Kyowa Kirin, Merck, Mundipharma, Novartis Pharmaceuticals, Ono Pharmaceutical, Sumitomo Dainippon Pharma, and Takeda Pharmaceuticals; and reports research support from AbbVie, Celgene, Chugai Pharmaceutical, Eisai, Janssen, Kyowa Kirin, Ono Pharmaceutical, Novartis Pharmaceuticals, and Takeda Pharmaceuticals. H.G. is an advisor or consultant for Celgene/Bristol Myers Squibb, Daiichi Sankyo, Novartis Pharmaceuticals, and Chugai Pharmaceutical; and reports honoraria from Celgene/Bristol Myers Squibb, Chugai Pharmaceutical, Janssen, Kyowa Kirin, Novartis Pharmaceuticals, SymBio, Takeda Pharmaceuticals, Eisai, and Daiichi Sankyo. A.A., C.L.-B., and S.C. are employees of Novartis Pharmaceuticals. M.R.B. is a consultant for Celgene, CRISPR Therapeutics, Juno Therapeutics, Kite Pharma, and Novartis Pharmaceuticals; reports research support from Novartis Pharmaceuticals; and reports participation on speakers bureaus for Celgene and Kite Pharma. The remaining authors declare no competing financial interests.

Correspondence: Ulrich Jaeger, Medical University of Vienna, Währinger Gürtel 18-20, 1090 Vienna, Austria; e-mail: ulrich.jaeger@meduniwien.ac.at.