Introduction Recent evidence showing benefit of blinatumomab (blina), a CD19xCD3, in upfront treatment of B-cell acute lymphocytic leukemia (B-ALL) broadly expands its use (Litzow et al., NEJM 2024). Due to risks of cytokine release syndrome (CRS) and neurotoxicity (NTX), the FDA label for blina specifies that patients (pts) with B-ALL should be hospitalized for a minimum of 9 days for cycle 1 (C1) (or 2 days if in complete remission (CR)) and 2 days for cycle 2 (C2). However, Gr3+ NTX and CRS are rare, particularly after C1 (Topp et al., Lancet Oncology 2015). As our ability to recognize and treat toxicities has improved, we assessed if blina can be safely administered outpatient (OP) during C1/C2.

Methods We conducted a retrospective cohort study of adult B-ALL pts treated with blina from 2012 to 2024 across 2 centers, Memorial Sloan Kettering Cancer Center (MSK) and Dana Farber Cancer Institute (DFCI). CRS and NTX were graded per ASTCT criteria (Lee et al., 2019). Hospital days were counted from the first dose of blina to 90 days after the last dose.

Results Overall, 307 pts (188 at MSK and 119 at DFCI) were included. 29% had active BM disease (≥5% blasts) at time of blina initiation.

Cycle 1: At DFCI, all pts received C1 inpatient, so this analysis focused on MSK pts. At MSK, 11 of 188 pts (6%) started C1 in the OP setting. Pts who started C1 as IP vs. OP did not differ in age (median 55 vs. 49 years, p = 0.8), proportion of Philadelphia positive (Ph+) ALL (34% vs, 45%, p = 0.4), CNS disease (3% vs. 0%, p > 0.9) at time of blina, or extramedullary disease (EMD) (20% vs. 17%, p > 0.9) at time of blina. No C1 OP pts had active BM disease, compared to 37% of C1 IP pts (p = 0.006). Six (55%) of C1 OP pts were MRD- at time of blina. C1 OP and C1 IP pts received a median of 2 vs. 2 cycles of blina (p = 0.7).

Of the 11 pts who started blina while OP, 3 (27%) required hospitalization for during the cycle (1 for NTX, 1 for pulmonary edema, 1 for subdural hemorrhage). C1 OP pts were less likely to have CRS in C1 than C1 IP pts (18% vs. 49%, p = 0.046) but rate of C1 NTX did not differ (9% vs. 13%, p > 0.9). No Gr3+ CRS/NTX was observed in the C1 OP group.

Healthcare utilization was numerically but not statistically lower for C1 OP vs. C2 IP pts: 6 vs. 14 median days of hospitalization per cycle (p = 0.4).

Cycle 2: Across both sites, 184 received a 2nd cycle; 101 (54%) started C2 IP and 83 (45%) OP. At DFCI, 73 pts (88%) started C2 IP and 10 (12%) started OP, while at MSK, 28 (28%) started IP and 73 (72%) OP. Pts who received C2 IP vs. OP did not differ statistically with regards to median age (54 vs. 56 years, p = 0.3) or rates of Ph+ disease (38% vs. 49%, p = 0.2), CNS1 status (96 vs. 100%, p = 0.3), EMD (7 vs. 12%, p = 0.2), or rate of active disease (29% vs. 22%, p = 0.5) at time of blina initiation. C2 OP pts received more cycles of blina than C2 IP pts (median 4 vs. 2, p < 0.001).

Of the 83 pts who started C2 OP, 11 (13%) were hospitalized during C2 (5 for infection, 2 for NTX, 2 for disease progression, 1 for possible bowel perforation, 1 for likely CRS). C2 OP pts were less likely to develop CRS at any time during blina compared to IP pts (43% vs 61%, p = 0.02) and rates of NTX did not differ statistically (16% vs. 8.4%, p = 0.12). Numerically more C2 OP pts from MSK had active disease compared to C2 OP pts at DFCI (25% vs 0%, p=0.11); despite this, no Gr3+ CRS/NTX was observed in C2 OP pts from either center, though Gr1 occurred numerically more frequently for the MSK C2 OP pts than DFCI (77% vs 40%, p=0.12).

Healthcare utilization was lower among C2 OP vs. C2 IP pts: 3 vs. 14 median days of hospitalization per cycle (p < 0.001). This remained statistically significant after stratification by center.

Conclusion In this large retrospective study across two major centers, blina was administered OP during C1 (to a small number of pts in CR) and C2 (to a significant number of pts with active disease or in CR before blina) without increased CRS or NTX compared to IP administration and was associated with lower healthcare utilization when given OP during C2. Our data suggests that C2 OP initiation may be a safe approach for most pts that may result in significant healthcare savings. As we gain experience with pts initiating blina in deep MRD- remissions and low total B-cell counts, our experience with C1 OP initiation of blina might help establish a safe pathway for this growing group of pts.

This content is only available as a PDF.
2025
Sign in via your Institution