Background: CAR-T cell therapy, while revolutionary for treating relapsed or refractory hematologic malignancies, faces economic and logistical hurdles in low and middle-income countries (LMIC) such as India (per-capita income of $2,100 compared to US $81,000). With treatment costs ranging from $300,000 to $500,000 in the US and total costs exceeding $700,000, affordability of CAR-T therapy, infrastructure limitations, and insufficient insurance coverage are significant barriers in India. In October 2023, India's Central Drugs Standard Control Organization approved Acalycabtagene Autoleucel (NexCAR19) to treat relapsed or refractory B-cell leukaemia (B-ALL) and lymphoma. The central question we addressed in this study was whether it is feasible to establish a CAR T cell therapy centre in resource limited setting to deliver the treatment safely and effectively.

Methods: A retrospective analysis was conducted from December 2023 to July 2024 at the haematology department of a tertiary care hospital in Kerala, South India. The department has a decade long running allogenic blood and marrow transplantation (BMT) program. Resources were limited, with most patients requiring to raise funds for their treatment through crowd funding. The core strategy involved leveraging the expertise of our existing BMT unit. This multidisciplinary team underwent comprehensive training in CAR therapy protocols and safety procedures. The existing infrastructure was optimized to meet the requirements for the therapy, ensuring compliance with necessary standards without new construction or major equipment purchases to optimize overall costs. Patients with relapsed or refractory B-cell malignancies who met regulator approved indications were selected. Leukapheresis was performed; CD3 and absolute lymphocyte counts were assessed. A cold chain was maintained for the leukapheresis product. Bridging therapy was given when deemed necessary. Patients underwent lymphodepletion five days before CAR-T infusion. Acalycabtagene Autoleucel was infused as per protocol, and patients were monitored for complications. Data on demographics, disease characteristics, adverse events, and responses were collected. Total costs, including CAR-T acquisition, administration, and inpatient care, were calculated. Variables were summarized by frequencies, or medians and ranges. Costs between patient subgroups were compared using t-tests.

Results: We treated 8 patients (6 males, 2 females), with a median age of 38 years (range 18-63). There were an equal number of patients with B-ALL and B cell lymphoma. Five received more than three lines of prior therapy. The median CD3 count at leukapheresis was 735 (206-1398) cells/ml. The median apheresis dose was 7.5 (5-13) x 10^6/kg Bridging treatment was administered to all except one. All patients were hospitalized 5 days prior to the date of infusion for lymphodepletion. Seven (87.5%) patients received fludarabine plus cyclophosphamide, while one (0.12%) received bendamustine. The time from apheresis to infusion was 28 days (range 11 to 41). At CAR-T cell infusion, 4 patients had active disease. Post-CAR T. B-cell aplasia was documented in five patients (62.5%). Complications included cytokine release syndrome (CRS) grade 1-2 in 7 (87.5%) patients, hemophagocytic lymphohistiocytosis (HLH) in 3 (37.5%), and sepsis in 2 (25%). Six (75%) patients achieved complete remission by day 28 post-infusion. The median treatment cost was $64,055 ($46,674-$81,571), with CAR-T product costs accounting for 74% of the total. B-cell lymphoma patients had a significantly lower cost (p < 0.01). Costs were higher for patients with complications. In particular, patients who needed treatment for HLH had significantly higher costs (p < 0.01). Five (62.5%) patients paid out of pocket, supported by crowdfunding campaigns from our public education initiatives, while three (37.5%) were covered by insurance or reimbursement.

Conclusions: Despite challenges such as high CAR-T costs, resource constraints, and patient affordability, the therapy was effectively implemented in an LMIC setting. Limitations included early-stage data, small sample sizes, and brief follow-up, yet the results highlight its potential for effective use in similar resource-constrained environments.

Disclosures

No relevant conflicts of interest to declare.

This content is only available as a PDF.
Sign in via your Institution