In this weeks NEJM there is an interesting article looking at the role of RT (or lack thereof) in preventing CNS relapse in newly diagnoses ALL.
This was a single arm trial which eliminated RT from the treatment protocol in all patients with newly diagnoses ALL. 498 were enrolled, however only 71 would have recieved prophylactic cranial irradiation (PCI) off protocol. These 71 were compared with 56 historical controls who did recieve PCI. Treatment intensity was risk based. "Triple intrathecal" therapy was used (consisting of MTX, cytarabine and hydrocoritsone).
5-yr EFS in 498 was 85.6% (95% CI, 79.9-91.3) and 5yr OS was 93.5% (95% CI, 89.8-97.2). 5yr isolated CNS relapse 2.7% (95% CI, 1.1-4.3), any CNS relapse was 3.9% (95% CI, 1.9-5.9). The 71 who did not have PCI had longer continuous complete remission than the 56 historical controls (P=0.04). Only one patient in the cohort that would have had PCI failed in the CNS.
CNS leukemia (CNS-3 status) or a traumatic LP with blasts and ≥1% residual disease after 6 weeks of remission induction had poorer EFS. CNS relapse risks included t(1;19), CNS involvement at diagnosis, and T-cell ALL. The authors however do not reccomend PCI in this cohort, as 90% of these patients would have recieved PCI unecessarily.
What the authors do not report (as it is not yet available) is on the long term neurocognitive and developemental consequences of intense intrathecal chemotherapy. The conclusions are predicated upon the assumption that this treatment will be an improvement over PCI (which may not be unreasonable), however only long term and in depth follow up will demonstrate this.