Factors associated with central nervous system relapse after b-acute lymphoblastic leukemia: A Children's oncology group report
Bibliographic record
Abstract
Abstract Introduction: While advancements in treatment of pediatric B-cell acute lymphoblastic leukemia (B-ALL), through incorporation of novel agents such as blinatumomab, have resulted in remarkable improvements in disease-free survival (Gupta, NEJM 2024), central nervous system (CNS) relapses remain a barrier. As immunotherapies which provide excellent medullary disease control but lack CNS penetration are incorporated into frontline treatment, there is a need to identify patients at risk of CNS relapse in whom deintensification of CNS directed therapy may be inappropriate, and whether these patients differ from those at risk of bone marrow (BM) relapse. Methods: Data from Children's Oncology Group (COG) trials for patients with de novo B-ALL (AALL0331, AALL0232, AALL0932 and AALL1131) were analyzed to examine factors associated with isolated CNS (iCNS), isolated BM (iBM) relapse and any CNS-involved relapse (CNS with any other site). Demographic and clinical characteristics were examined. Cumulative incidence of iCNS, iBM and any CNS-involved relapse was evaluated, treating death, induction failure, subsequent neoplasm and other relapses as competing events. Proportional cause-specific hazards models were used to estimate the association of covariates with the hazard of iCNS, iBM and any CNS relapse. Results: Among the 21,830 patients enrolled across the four trials between 01/2004 and 08/2019 (54.6% male; median age: 4 years [y] [range, 1-30y]), 2,238 (10.3%) patients relapsed (473 [21.2%] iCNS, 1,341 [60.1%] iBM, and 671 [30.0%] any CNS). The 5-year cumulative incidence of relapse among all patients was 9.5%±0.2% and for iCNS, iBM and any CNS-involved relapse was 2.2%±0.1%, 5.6%±0.2% and 3.2%±0.1%, respectively. Median time to relapse was earlier for iCNS compared to iBM: 24.6 vs. 38 months. Multivariable analysis demonstrated that age at diagnosis was not associated with iCNS relapse (10-15y: Hazard ratio [HR]=1.2, 95% confidence interval [CI]=0.9-1.5; ≥16y: HR=1.3, 95%CI=0.9-1.7; ref=1-9y) but was associated with iBM relapse (10-15y: HR=1.6, 95%CI=1.4-1.8; ≥16y: HR=1.7, 95%CI=1.5-2.1; ref=1-9y). Compared to males, females were at a lower hazard of iCNS relapse (HR=0.5, 95%CI=0.4-0.6) but greater hazard of iBM relapse (HR=1.2, 95%CI=1.0-1.3). Compared to Non-Hispanic White patients, Non-Hispanic Black race/ethnicity was associated with greater hazard of both iCNS (HR=1.8, 95%CI=1.3-2.5) and iBM relapse (HR=1.7, 95%CI=1.4-2.1) with similar magnitudes of risk, whereas Hispanic ethnicity was associated with greater hazard of iBM relapse only (HR=1.2, 95%CI=1.0-1.3). The findings for any CNS-involved relapse were similar to iCNS relapse, with the addition of Hispanic ethnicity associated with higher risk (HR=1.2, 95%CI=1.0-1.5). Favorable blast genetics were associated with reduced hazard of both iCNS and iBM relapse. Unfavorable blast genetics were associated with greater hazard of iBM relapse (hypodiploidy: HR=4.6, 95%CI=3.6-5.7; BCR::ABL1: HR=1.9, 95%CI=1.5-2.5; ref=neutral cytogenetics) but not iCNS relapse (hypodiploidy: HR=0.3, 95%CI=0.1-1.1; BCR::ABL1: HR=1.3, 95%CI=0.8-2.2; ref=neutral cytogenetics), with the exception of intrachromosomal amplification of chromosome 21 (iAMP21), which was associated with both (iBM relapse: HR=1.8, 95%CI=1.5-2.3; iCNS relapse: HR=1.8, 95%CI=1.2-2.7). CNS status at diagnosis was associated with greater hazard of iCNS relapse (CNS2: HR=2.2, 95%CI=1.7-2.7; CNS3: HR=2.5, 95%CI=1.5-4.0; ref=CNS1) as well as any CNS relapse but not iBM relapse (CNS2: HR=1.1, 95%CI=0.9-1.2; CNS3: HR=1.1, 95%CI=0.8-1.7; ref=CNS1). EOI MRD was also associated with both, however magnitude of association was substantially higher for iBM relapse (0.01%-0.099%: HR=3.0, 95%CI=2.6-3.5; 0.1%-0.99%: HR=4.2, 95%CI=3.6-4.8; ≥1.0%: HR=4.5, 95%CI=3.7-5.4; ref=<0.01%) than iCNS relapse (0.01%-0.099%: HR=1.5, 95%CI=1.2-2.0; 0.1%-0.99%: HR=1.5, 95%CI=1.1-2.0; ≥1.0%: HR=1.6, 95%CI=1.0-2.3; ref=<0.01%). The findings for any CNS-involved relapse were similar to iCNS. Conclusions: To our knowledge, this is the largest analysis of factors associated with CNS relapse to date. We find that predictors of CNS relapse differ from those of BM relapse, with implications for future trials testing treatment strategies which rely on immunotherapy to de-intensify traditional chemotherapy. Novel and more powerful predictors of CNS relapses and more effective CNS-directed therapies are needed to fully optimize the potential of immunotherapy.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.001 | 0.003 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.001 | 0.002 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.000 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.000 | 0.001 |
| Insufficient payload (model declined to judge) | 0.002 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".