Relapsed/refractory B-cell acute lymphoblastic leukemia (R/R B-ALL) is associated with dismal clinical outcomes, with long-term cure rates of 10% with conventional therapies [1]. Inotuzumab Ozogamicin (InO) is a humanized anti-CD22 monoclonal antibody-drug conjugate that has shown significant activity in adult patients with R/R B-ALL as demonstrated in the phase III INO-VATE trial [2]. There is limited real-world data on the use of InO for R/R B-ALL with reports mainly from the global north [3,4,5]. There is reasonable consensus even in a low-middle income country setting for the use of InO [6] but limited data on clinical outcomes [7], especially in the setting of dose modifications due to financial constraints.

Following Institutional Review Board Approval, a retrospective study was conducted to analyze the data of patients with R/R B-ALL at our center from January 2020 to December 2024. During the study period, 127 patients had R/R B-ALL at our center. Amongst these, 83 patients were treated at our center. Of these, 49 (median age 24 years, range 6–68) received InO for refractory (n = 9) or relapsed (n = 40) B-ALL. The details of these patients are shown in Table 1. Cytogenetics was available for 46 patients and of these, 6 patients had Philadelphia chromosome positive (Ph + ) B-ALL. Due to cost considerations, some patients received fixed doses of 1 mg vial of InO on either days 1, 8, and 15 (n = 25) or days 1 and 15 (n = 2). The remainder of the patients received the dose as per body surface area (n = 22). The median dose of InO was 1.72 (range 1.23–2.1) mg/m2/cycle amongst the responders versus 1.54 (range 0.61–1.82) mg/m2/cycle amongst the non-responders (p value 0.42). Among the 44 patients for whom the CD22 data was available, the median CD22 expression was 98.3 (16.9–99.9)% among the responders and 94.92 (1.13–99.8)% among the non-responders (p value 0.46).

Table 1 Baseline characteristics and outcomes (n = 49).
Full size table

Twenty-eight (57.14%) patients had a response (defined as CR/CRi as per ELN 2024 guidelines [8]), amongst whom 20 (71.4%) were MRD negative. Among the responders, 19 (67.8%) underwent a subsequent transplant or cellular therapy (18 patients underwent an allogeneic stem cell transplant (conditioning details are mentioned in Supplementary Table 1) and 1 patient underwent CAR-T cell therapy).

In our cohort, 15 (30.6%) patients developed febrile neutropenia, 10 of the 49 patients who received InO (20.4%) experienced hepatic sinusoidal obstruction syndrome (SOS) (see Supplementary Table 2), of which 9 (i.e., 50% of the transplant cohort) occurred post-transplant despite the use of prophylactic ursodiol in all patients in the peri-transplant period. The median time from last InO to transplant was 45 (30–120) days. The median dose of InO amongst those who had severe SOS (n = 8) was 1.79 (range 1.51–2) mg/m2/cycle when compared to 1.61 (range = 0.75–2.1) mg/m2/cycle amongst patients who did not. There was no statistically significant correlation between the occurrence of SOS and the number of doses of InO (p value 0.42), cumulative dose of InO (p value 0.42) or time interval between last dose of InO and transplant (p value 0.47). Of the 10 who developed SOS (see Supplementary Table 2), 8 had severe/very severe SOS (as defined by the EBMT criteria [9]). Five had resolution of SOS, and the median OS after recovery from SOS was 14 months (range, 2–27 months).

The 1-year overall survival for the entire cohort was 51.7% ± 8.1%, with the 1-year event-free survival (event defined as death or relapse) being 32.4% ± 8%. The 1-year overall survival for patients undergoing transplant/CAR T was 68.4% ± 10.7% (vis-à-vis 42.3% ± 20.6% for patients who were in remission, but did not undergo transplant/CAR T, log rank p 0.129), as shown in Fig. 1. The median follow-up for survivors in the transplant/CAR-cohort was 20.1 months (range 2–33.5). Supplementary Figs. 1, 2 show the OS and EFS comparison between primary refractory versus relapsed and first-line versus later-line salvage respectively. Supplementary Fig. 3 shows the subsequent therapies following response and non-response to InO salvage.

Fig. 1
figure 1

A Overall survival (OS) and event-free survival (EFS) of the entire cohort (n = 49). Event was defined as death or relapse. The 1-year OS for the entire cohort was 51.7% ± 8.1% and the 1-year EFS was 32.4% ± 8%. B 1-year OS for the patients undergoing a subsequent transplant/CAR T was 68.4% ± 10.7% (vis-à-vis 42.3% ± 20.6% for patients who had response and did not undergo transplant/CAR T, log rank p 0.192). 1-year EFS for the patients undergoing a subsequent transplant/CAR T was 47.1% ± 12.3% (vis-à-vis 33.3% ± 19.2% for patients who had response and did not undergo transplant/CAR T, log rank p 0.743).

Full size image

In a large real-world analysis from the Center for International Blood and Marrow Transplant Research (CIBMTR) [10], adult patients with ALL who received InO before undergoing transplant were evaluated for post-transplant hepatic toxicity. The study reported an incidence of SOS of 14% within 100 days post-transplant, with rates rising to about 18% in R/R ALL. In a study by Senapati et al. [11], which evaluated liver toxicity risk and SOS incidence among adults with ALL receiving InO-based regimens, majority of patients were treated with fractionated InO dosing schedules as part of combination chemotherapy protocols such as mini-Hyper-CVD with InO. This dosing approach resulted in lower SOS rates compared with historical single-dose or higher-intensity regimens reported in earlier trials (e.g., INO-VATE). The incidence of SOS was 9.8% overall and 17.3% among those proceeding to transplant, which was lower than the historical incidence of 20–30% observed with non-fractionated, higher-dose InO exposure. These findings suggest that dividing the total InO dose into smaller, fractionated infusions within each cycle, and combining it with less hepatotoxic chemotherapy, mitigates peak calicheamicin exposure and endothelial injury, thereby reducing the risk of SOS while maintaining antileukemic efficacy.

Our analysis is limited by the relatively small sample size and short follow-up. We conclude that InO salvage in R/R B-ALL is associated with reasonable remission rates, even when less than recommended doses were used, enabling a subsequent transplant or cell therapy in these patients in our real-world cohort. However, strategies like the use of fractionated InO along with chemotherapy, preferably as first salvage, to reduce post-transplant SOS rates and strategies to reduce post-transplant relapses need to be explored.