Pediatric Chemotherapy Desensitization: Breakthrough Reaction Risks
Key Takeaways
- At a Bangkok pediatric center, children undergoing chemotherapeutic rapid drug desensitization had breakthrough reactions in 15.9% of procedures, and most of those reactions were mild.
- Among pediatric chemotherapeutic rapid drug desensitization procedures complicated by a breakthrough reaction, about two-thirds still reached the full therapeutic dose after treatment.
- E. coli-asparaginase desensitization and a higher starting dose were associated with more breakthrough reactions during pediatric chemotherapeutic rapid drug desensitization.
- In this cohort, all CRRs had breakthrough reactions, whereas IRRs and mixed-type reactions did not, and later desensitization cycles were associated with lower risk.
An ambispective observational cross-sectional analysis at Siriraj Hospital Mahidol University in Bangkok, Thailand, included patients aged 18 years or younger who had immediate hypersensitivity reactions (HSRs) within 1 to 6 hours of chemotherapeutic exposure and then underwent rapid drug desensitization (RDD) between January 2010 and April 2025. The cohort comprised 38 children, 52 suspected initial HSRs, and 182 RDD procedures. Protocols were individualized, typically starting at 1/1,000,000 to 1/10,000 of the therapeutic dose, escalating every 15 minutes with multiple infusion bags in a pediatric intensive care unit or similarly monitored setting under pediatric allergist supervision; phenotypes included type I HSRs, infusion-related reactions (IRRs), and cytokine-release reactions (CRRs). The analysis focused on breakthrough reaction (BTR) frequency, procedure completion, reaction phenotype, and factors associated with risk.
Acute leukemia accounted for 47.4% of the cohort, and E. coli-asparaginase, vincristine, and methotrexate were the leading culprit agents at 26.9%, 24.7%, and 13.7%, respectively. BTRs occurred in 29 of 182 RDD procedures (15.9%), and 72.4% were mild, with the remainder moderate or severe. Type I HSRs accounted for most initial reactions, all four CRR cases experienced BTRs, and no BTRs were observed in IRR or mixed-type reactions in this cohort. These patterns helped identify which procedures were more likely to react during desensitization.
BTRs were more frequent with E. coli-asparaginase than with other agents (55.2% vs 21.6%, p < .001). They were also more frequent when the initial starting ratio was 1:10,000 or less (24.1% vs 5.2%, p = .003), and they clustered earlier in the desensitization course, with median cycle 1 (IQR 1-2) versus cycle 3 (IQR 2-6), p < .001. In the multivariable analysis of breakthrough-reaction risk factors, E. coli-asparaginase exposure carried an adjusted OR of 3.87 (95% CI 1.45-10.39; p = .007), an initial starting ratio of 1:10,000 or less had an adjusted OR of 3.51 (95% CI 1.02-12.18; p = .048), and each additional cycle had an adjusted OR of 0.47 (95% CI 0.31-0.72; p < .001). Among procedures complicated by a BTR, 19 of 29 (65.5%) were completed after treatment and 10 of 29 (34.5%) were terminated before the target dose. Incomplete procedures more often involved respiratory symptoms, E. coli-asparaginase, and intramuscular adrenaline use.
This single-center cohort was drawn mostly from retrospective case identification, and RDD protocols were individualized rather than standardized, making protocol-related variables harder to interpret and limiting causal inference. Phenotype assignment relied mainly on clinical history, paired tryptase was available only in a subset, and interleukin-6, anti-drug antibodies, and other immunologic biomarkers were not systematically measured, which constrained mechanistic interpretation. Culprit drugs were confirmed in 20 of 38 patients, corresponding to 24 of 52 suspected drugs, and serum tryptase data were available for 11 patients. Within those bounds, the higher BTR burden in this cohort was linked chiefly to E. coli-asparaginase exposure, a higher initial starting dose, and earlier desensitization cycles.
The authors reported that pediatric chemotherapeutic RDD generally preserved access to treatment in most children with immediate HSRs, even though recurrent reactions during desensitization remained an interruption point. BTRs were usually mild but still prevented full target dosing in a meaningful minority of affected procedures, and the main factors associated with those reactions were E. coli-asparaginase, higher initial dosing, and earlier desensitization cycles.
Clinician Questions
How often did breakthrough reactions occur during pediatric chemotherapy rapid drug desensitization?
In pediatric chemotherapy rapid drug desensitization, breakthrough reactions occurred in 29 of 182 procedures, or 15.9%, and 72.4% of those breakthrough reactions were mild.
Which factors were independently associated with breakthrough reactions during pediatric chemotherapeutic desensitization?
During pediatric chemotherapeutic desensitization, E. coli-asparaginase exposure was associated with higher odds of breakthrough reactions with an adjusted OR of 3.87 (95% CI 1.45-10.39), and an initial starting ratio of 1:10,000 or less was also associated with higher odds with an adjusted OR of 3.51 (95% CI 1.02-12.18), while each additional desensitization cycle was associated with lower odds with an adjusted OR of 0.47 (95% CI 0.31-0.72).
How often were pediatric desensitization procedures completed after a breakthrough reaction?
Among 29 pediatric chemotherapeutic desensitization procedures complicated by breakthrough reactions, 19 were completed after treatment and 10 were terminated before the target dose, corresponding to 65.5% completion and 34.5% termination.
Did breakthrough reactions differ by reaction phenotype during pediatric chemotherapy desensitization?
In this pediatric chemotherapy desensitization cohort, 42.1% of patients with type I hypersensitivity reactions experienced breakthrough reactions, no breakthrough reactions were observed among infusion-related or mixed-type reactions, and all four patients with cytokine-release reactions experienced breakthrough reactions.