The classification study POG 9900 used NCI age and WBC criteria for standard and high risk ALL, at diagnosis, to assign patients to a dexamethasone (D) based 3-drug or a prednisone based 4-drug induction. Refined age and WBC criteria and blast cytogenetics were used at end induction to assign patients to low, standard or high risk post-induction therapy. An objective of this study was to study the correlation between outcome and flow cytometric determination of MRD, measured in peripheral blood (PB) on day 8 and in marrow on day 29 of induction. Though each component of the 3-drug induction had been safely included in previously published pediatric protocols, the identical dosing and scheduling pattern of these agents, as used in this protocol, had not been previously tested. Unexpected, excessive toxicity associated with the original 3-drug induction necessitated changes in the induction therapy. This created an opportunity to assess the impact of changes in therapy on MRD. The original induction included D-6 mg/m 2 /day x 28 days, vincristine-1.5 mg/m 2 weekly x 4, E Coli asparaginase (A)-10,000 IU/m 2 /dose on days 2, 5, 8, 12, 15, and 19 and intrathecal methotrexate (ITM) on days 1 and 8. The 0.87% (n = 1263) induction mortality rate associated with the 3-drug induction was significantly higher than that associated with previous POG and CCG protocols, mandating therapy amendment, changing the day 1 ITM to IT cytarabine (ITA) and the 6 doses of A to 1 dose of pegasparaginase (PEG), given on either day 4, 5 or 6. This modified induction regimen was identical to that used in parallel in CCG 1991. There were 2 subsequent induction deaths ( 0.38%, n=527). MRD values before and after the change in induction therapy were: The increase in the percentage of MRD positive patients on day 8, PstAm, likely reflects the change from ITM to ITA, reflecting a loss of the systemic anti-leukemic effect of ITM. Given the concomitant change from A to PEG, this result is surprising because CCG 1962 had demonstrated an increase in the proportion of patients achieving an M1 marrow on day 7 with PEG vs A. However, A was started on day 2 PreAm vs the administration of PEG on day 4, 5 or 6, PstAm. The later administration of PEG may have impacted the D8 early response. Of interest, however, despite the higher prevalence of day 8 MRD in the PstAm patients, by day 29 a significantly lower percentage of the patients treated PstAm were MRD positive (RR=0.76). We report separately that MRD >0.01% at end induction identified a patient subgroup with poor outcome, suggesting that PEG may improve long-term outcome as compared to A. Thus, what appear to be relatively minor changes (ITA vs ITM, A vs PEG); can have substantial effects on induction mortality and early response measures. This simultaneously highlights the narrow therapeutic index associated with cytotoxic chemotherapy and suggests that early MRD measures may be useful as a surrogate for long-term EFS.