Intensive but tender care for infant ALL

In this issue of Blood, Tomizawa et al report remarkably improved outcomes for 90 infants with acute lymphoblastic leukemia (ALL) by providing excellent supportive care while receiving intensive chemotherapy and the use of hematopoietic cell transplant (HCT) for specific high-risk groups.¹ 

Infant ALL is a rare (20 cases per million per year) but aggressive leukemia characterized by KMT2A rearrangements (in ∼75% of cases) and dismal outcomes (see table).^2-4  Two large international collaborative randomized trials^5,6  were performed to improve the prospects for infants, using standard vs more intensive therapy before maintenance treatment (Interfant-99 study) or myeloid- vs lymphoid-type consolidation therapy (Interfant-06 study). However, the event-free survival (47.0% and 46.1%) and the overall survival (55.3% and 58.2%) at 4 and 6 years, respectively, for the 2 studies indicated that neither treatment modification had a substantial impact on prognosis.

Infants are physically and immunologically immature and are truly vulnerable to treatment-related complications, especially infection, with ∼10% dying from treatment-related complications.^3-6  The benefit of HCT in infants with ALL is still controversial.^2,4,5,7  A retrospective international study²  and the CCG1953 study⁴  showed that HCT failed to improve outcome in infants with ALL, with the possible exception of the subgroups defined by KMT2A rearrangement and age <6 months plus either a white blood cell (WBC) count ≥300 × 10⁹/L or a poor response to prednisone prephase in the Interfant-99 study.⁷ 

Tomizawa et al achieved an outstanding 5-year event-free survival of 70.9% and an overall survival of 85.0% in their MLL-10 study, despite using the same induction regimen and the same postinduction drugs as were used in the Interfant-99 study.⁵  How did this study group achieve these stellar results?

First, the MLL-10 study provided extended supportive care. In the Interfant-06 study, the overall survival rate was significantly better among patients treated by the original participating national study groups in Western Europe and North America, compared with those treated by the newly participating study groups in the other regions (62.1% vs 49.7%).⁶  The worse outcome was attributed to higher rates of mortality during induction, refractory leukemia, and death in remission in the less experienced groups that joined the trial later, which was thought to reflect fundamental differences in supportive care.⁶  In this regard, the Japanese health care system allows continuous hospitalization from diagnosis until the completion of induction, early consolidation, reinduction, and late consolidation treatment (up to weeks 23 to 26 of therapy). This unique strategy affords a very protective patient-care environment, close monitoring, and immediate use of critical supportive-care measures, including prophylactic and empirical use of antimicrobial and antifungal agents and granulocyte colony-stimulating factor.

Second, better supportive care permits early and more robust treatment intensification. Because infant ALL arises from an immature B-lymphoid progenitor (often with CD10 negativity) with coexpression of myeloid antigens, cytarabine is considered a key drug in this disease, just as in acute myeloid leukemia.⁸  Hence, high-dose cytarabine with asparaginase is commonly used in infant ALL. While the Interfant-06 and the COG AALL0631 studies used this combination at later phases of chemotherapy,^6,8  the MLL-10 protocol introduced it in the early consolidation phase for intermediate- or high-risk patients.¹  Moreover, because of their different physiologic features compared with those of older children (eg, body composition, drug-plasma protein binding, and cytochrome p450 activity), infants often received reduced dosages of chemotherapy. In the MLL-10 study, more stringent criteria for age-based chemotherapy dose reduction were applied.¹  These decisions could have resulted in better clearance of minimal residual disease at the end of the early consolidation phase, which proved to be the most important prognostic factor in the MLL-10 study. Remarkably, Tomizawa et al report no early deaths during induction or early consolidation, and only 2 deaths in complete remission in the later phase of their study.

Finally, the high-risk criteria used by the MLL-10 study (KMT2A rearrangement with either age <6 months or central nervous system involvement [≥5/µL WBC with ALL morphology]) were broader than those in the Interfant-06 study, resulting in 56 (62.2%) patients classified to have high-risk ALL.¹  HCT was performed in 43 patients, representing 87.8% of the 49 high-risk infants in remission and 47.8% of the entire cohort; 67.4% of the transplanted infants were alive in remission. In the Interfant-06 study, more stringent high-risk criteria were used, and there were many “early events” (n = 54), resulting in only 76 (53.1%) of the 143 high-risk patients in remission and 111 (17.1%) of the entire cohort of 651 infants receiving HCT. The transplanted high-risk patients had 4-year disease-free survival of 44.0%.⁶  Had these patients received an intensified early phase of chemotherapy with aggressive supportive care, they would likely have achieved better leukemia clearance and physical condition prior to HCT, both of which are critical for a successful outcome. It may also be important that the MLL-10 was performed only in Japan, a nation with a relatively homogenous population and less HLA disparity than typically found in other countries.⁹  These factors might all have contributed to a lower rate of transplant-related death (2.3% vs 14.4%) and better disease-free survival than reported for the Interfant-06 study.⁶ 

Because survivors of infant ALL can look forward to many decades of productive life, the possibility of therapy-related complications, particularly late effects, pose a major concern. Effective agents are needed for patients with KMT2A rearrangements, specifically those with persistent minimal residual disease, to boost current survival rates. Targeting unique molecular characteristics of KMT2A-rearranged ALL with DOT1L, bromodomain, menin, and BCL-2 inhibitors has shown promise in preclinical or early clinical studies,⁸  as have immunotherapeutic approaches with the bispecific antibody blinatumomab and autologous and allogeneic chimeric antigen receptor T-cell therapy, albeit the possibility of lineage switch to acute myeloid leukemia in cases with KMT2A rearrangement.^8,10 

In summary, the treatment approach taken by Tomizawa et al is important because it illustrates the yin and yang of a successful protocol for infant ALL. That is, it may be possible to improve outcome by judicious intensification of therapy, but only if the hazards of aggressive treatment are tempered with extended and comprehensive supportive measures, or “tender care.”