Risk of hepatitis B reactivation is controllable in patients with B‐cell lymphoma receiving anti‐CD19 CAR T cell therapy

Hepatitis B virus (HBV) reactivation is a well-recognised complication in lymphoma patients with concomitant HBV infection, who are receiving B cell-depleting agents, such as rituximab.1, 2 Anti-CD19 chimeric antigen receptor (CAR) T cell is a promising therapy for patients with relapsed or refractory B-cell lymphoma. As B-cell aplasia is a common adverse event after anti-CD19 CAR T cell therapy,3, 4 patients with HBV infection receiving anti-CD19 CAR T cell therapy may have the same high risk of HBV reactivation as other B cell-depleting therapies. To date, reports of HBV reactivation after anti-CD19 CAR T cell therapy are rare. We performed a post-hoc analysis of two prospective clinical trials (NCT03029338, ChiCTR1900025419) of CNCT19 (autologous second-generation anti-CD19 CAR T cells using 4-1BB as a co-stimulatory domain provided by Juventas, Tianjin, China) therapy in relapsed or refractory B-cell lymphoma patients, and aimed at exploring the risk of HBV reactivation in patients with concomitant HBV infection, receiving CNCT19 therapy. Between June 2017 and June 2019, 17 relapsed or refractory B-cell lymphoma patients with concomitant chronic [defined as positive hepatitis B surface antigen (HBsAg), n = 6] or resolved [defined as negative HBsAg but positive antibody to hepatitis B core antigen (anti-HBc), n = 11] HBV infection were included in our study. Among them, 14 patients received CNCT19 therapy alone, and three received CNCT19 following high-dose chemotherapy and autologous stem cell transplantation (HDT/ASCT). The HBV-DNA levels of all 17 patients were lower than 1,000 iu/ml (normal limit of the test) at baseline. HBV reactivation was defined as an elevation of HBV-DNA levels to more than 1,000 iu/ml and/or HBsAg reverse seroconversion in HBsAg-negative patients. For patients receiving CNCT19 therapy alone, a lymphodepleting chemotherapy consisting of fludarabine and cyclophosphamide was given, and CNCT19 (median dose, 2·5 × 106/kg, range 1·7–4 × 106/kg) was administered in single-dose or split-dose infusion on day 0 or day 0,1. For patients who received CNCT19 infusion following HDT/ASCT, a conditioning regimen of GBC (gemcitabine, busulfan and cyclophosphamide) was administered, and a fixed dose of CNCT19 (2·0×106/kg) was infused on day +2, +3 or +4 following stem cell infusion. All patients received rituximab during their prior therapy. The median interval between the last dose of rituximab and CNCT19 infusion was three months (range, 1‒24 months). Table I illustrates the baseline demographics, disease characteristics and HBV serological status of each patient. At the time of screening, all six patients with chronic HBV infection were on prophylactic entecavir, while only two of the 11 patients with resolved HBV infection were on prophylactic nucleos(t)ide therapy (NAT) (Table I). During and after CNCT19 therapy, all six patients with chronic HBV infection continued to receive prophylactic entecavir, and five of 11 patients with resolved HBV infection received prophylactic NAT. With a median follow-up of 10 months (range, 1–30 months) from CNCT19 infusion, none of these patients developed HBV reactivation. One patient with chronic HBV infection self-discontinued entecavir at nine months after CNCT19 infusion following HDT/ASCT and did not experience HBV reactivation by the time of the last follow-up (24 months after CNCT19 infusion). The other five patients with chronic HBV infection all remained on continuous prophylactic entecavir (median, 10 months, range, 8–29 months). Of the five patients with resolved HBV infection who received prophylactic NAT during CNCT19 therapy, the median duration of antiviral prophylaxis was two months (range, 1–10 months). Four patients stopped prophylactic NAT at 1, 2, 2 and 6 months after CNCT19 infusion, respectively. Fig 1 lists the duration of prophylactic NAT of each patient. Among the 11 patients with resolved HBV infection, 10 (90·9%) had positive anti-HBs, and six (54·5%) had anti-HBs titers higher than 56·48 miu/ml. After CNCT19 therapy with or without HDT/ASCT, the reduction of anti-HBs titers was observed in two patients (Table SI). The anti-HBc values of 11 patients were all lower than 10 S/CO at baseline, with a median anti-HBc value of 2·89 S/CO (range, 1·14–8·93 S/CO) (Table I). We detected CD19-positive B cells in peripheral blood using flow cytometry and found that all patients developed B-cell aplasia, except for two patients who had neoplastic CD19-positive B cells in blood and exhibited no response to CNCT19 therapy. The median time of B cell recovery post-CNCT19 infusion among the 12 assessable patients was seven months. Wei et al. recently reported a case of a B-cell lymphoma patient with a chronic HBV infection who experienced HBV reactivation and fulminant liver failure after self-discontinuing prophylactic entecavir at one month after anti-CD19 and anti-CD22 CAR T cell sequential infusion.5 However, most of our patients with chronic HBV infection received prolonged prophylactic entecavir, and none developed HBV reactivation, which indicated that prophylactic NAT could protect this subgroup of patients against HBV reactivation during and after CNCT19 therapy. Several researchers have suggested that positive anti-HBs can protect patients with concomitant HBV infection against HBV reactivation after receiving rituximab-containing therapy,1, 6, 7 and Yang et al. have defined an anti-HBs titer of 56·48 iu/ml as a cut-off value for predicting HBV reactivation.6 Of the 11 patients with resolved HBV infection in our study, 90·9% had positive anti-HBs at baseline, while 54·5% had anti-HBs titers higher than 56·48 iu/ml, which might be an underlying reason for the low risk of HBV reactivation in our cohort. The anti-HBc levels can reflect the residual HBV-DNA load, and low anti-HBc levels (<10 S/CO) at baseline may also be a good marker for predicting HBV reactivation.8 All patients with resolved HBV infection in our study had low anti-HBc values with a median baseline anti-HBc value of 2·89 S/CO, which might be another reason for the low risk of HBV reactivation among our patients. Collectively, anti-CD19 CAR T cell therapy could be safely administered in B-cell lymphoma patients with concomitant HBV infection. As the duration of B-cell aplasia post-CNCT19 infusion is similar to that of patients treated with rituximab, we suggested that the same strategy of antiviral prophylaxis as that in patients receiving rituximab-containing therapy should be administered in patients treated with CNCT19 cells. Given that the risk of HBV reactivation seemed low among HBsAg-negative/anti-HBc-positive patients who had positive anti-HBs and/or low anti-HBc values in our study, the strategy of HBV-DNA monitoring-guided preemptive NAT might be explored in this cohort in the future. The authors thank the patients who participated in the study, their families, friends, and caregivers, and all clinical and laboratory members of the Institute of Hematology & Blood Diseases Hospital for support and help with this study. This study was supported by the Chinese Academy of Medical Sciences (CAMS) Innovation Fund for Medical Sciences (CAMS-2016-I2M-3-013). JW served as a consultant for AbbVie, and received a grant from Celgene. The remaining authors declare no competing financial interests. WL and WH contributed to the conception and design of the study, treatment of patients, collection and assembly of data, data analysis and interpretation, and manuscript writing; MW and QR contributed to CAR and product design; RL, JL, SD, SY and HL contributed to enrollment and treatment of patients; YW contributed to the conception and design of the study; YX contributed to the measurement of CAR T cell proliferation in vivo; LL contributed to the provision of essential reagents; LQ contributed to the conception and design of the study; DZ served as principal investigator and contributed to the conception and design of the study; JW contributed to the CAR and product design and the conception and design of the study. Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.