A large randomized trial has found that patients with small-cell lung cancer who underwent regular brain-MRI monitoring without preventive brain radiation had better cognitive failure-free survival than those who received both monitoring and prophylactic cranial irradiation. The results, presented Sept. 13 at the International Association for the Study of Lung Cancer World Conference on Lung Cancer in Seoul, could reshape a long-debated treatment decision for patients who have completed their initial cancer therapy and have no brain metastases visible on MRI.
The study does not yet settle whether the two approaches produce the same overall survival. In the preliminary analysis of the SWOG S1827 MAVERICK trial, survival did not appear meaningfully different, but the confidence interval allowed for either benefit or harm and the planned final analysis has not been reached. The distinction is important because preventive cranial irradiation, or PCI, has historically been used to reduce the chance that small-cell lung cancer will spread to the brain, despite concerns about cognitive effects and other toxicities.
What the trial tested
The international phase III trial enrolled 304 people with either limited-stage or extensive-stage small-cell lung cancer between January 2020 and December 2025. All had finished their initial treatment and had an MRI showing no brain metastases before they entered the study, according to the IASLC conference release on the MAVERICK results.
Participants were randomly assigned to one of two strategies. One group received scheduled MRI surveillance alone. The other received the same monitoring plus PCI, delivered as 25 gray of radiation over 10 treatment sessions. Both groups had brain MRIs every three months during the first year and every six months during the second year. They underwent cognitive testing on the same timetable, using assessments of verbal learning, word generation and processing speed.
That schedule is integral to the result, not a minor detail. MRI surveillance is intended to identify brain metastases early enough for treatment if they emerge, while sparing patients radiation that is given before any detectable brain disease. The study therefore tested an organized follow-up strategy, not simply the omission of radiation. Its findings do not automatically apply to people who cannot receive frequent, high-quality MRI follow-up or to patients whose scans already show brain metastases.
Cognition and serious adverse events favored surveillance alone
After a median follow-up of 22 months among patients still alive, MRI surveillance alone was associated with better cognitive failure-free survival, the trial’s primary endpoint. The reported hazard ratio was 0.60, with a 90% confidence interval of 0.46 to 0.78 and a p value of 0.001. In practical terms, patients assigned to surveillance alone had a lower relative rate of reaching the trial’s cognitive-failure endpoint during follow-up than patients assigned to PCI plus surveillance.
Because MAVERICK randomly assigned the treatment strategies, the comparison is stronger evidence of a treatment-related difference than an observational comparison would be. But the conference report did not provide absolute cognitive-failure rates, full quality-of-life findings, treatment-adherence data or complete baseline and adverse-event tables. Those details will be important for judging the size of the benefit for individual patients and whether it was similar across demographic and clinical groups.
Serious treatment-related adverse events were also less frequent in the MRI-alone group: 0.8%, compared with 7.9% in the PCI group. One grade 5 encephalopathy event, meaning a treatment-related death, occurred in the PCI group. The release reported that the benefit of surveillance alone appeared consistent in both limited-stage and extensive-stage disease and regardless of immunotherapy use, but it did not include subgroup-specific estimates or formal interaction-test results.
Survival question remains open
The early overall-survival analysis was based on 128 deaths. It produced a hazard ratio of 0.90 for MRI surveillance alone compared with PCI plus surveillance, with a 90% confidence interval of 0.67 to 1.20. Since that interval includes 1.0 and spans possibilities in both directions, it cannot establish either a survival advantage for surveillance alone or definitive survival equivalence between the approaches.
Investigators plan the final survival analysis after 190 deaths. The reported analysis also found no statistically significant difference in brain-metastasis-free survival: the hazard ratio was 1.25, with a 90% confidence interval of 0.95 to 1.66. That result should not be read as showing that MRI monitoring eliminates the risk of brain metastases. Rather, it leaves uncertainty about whether the two strategies differ in the time until brain metastases occur under the trial’s monitoring and treatment framework.
The presenter, Chad Rusthoven of the University of Colorado School of Medicine, said the results support MRI surveillance alone as a standard-care approach. That is an interpretation of the newly presented data, however, not documentation that professional guidelines have already changed. The currently available account is a conference release rather than a full peer-reviewed publication, and the survival follow-up is incomplete.
Why preventive radiation has remained contested
Small-cell lung cancer is an aggressive disease with a substantial risk of spreading to the brain. PCI became established in part because earlier studies found it could lower the occurrence of brain metastases in selected patients. Yet the balance of benefit and harm has been reconsidered as MRI has become more available and clinicians have been able to find asymptomatic brain metastases earlier through routine imaging.
A peer-reviewed review of PCI and MRI surveillance describes conflicting evidence in the MRI era, particularly for extensive-stage disease. Earlier trials did not always require the brain imaging that is now common, making it difficult to separate the value of preventive radiation from the value of earlier detection and treatment of metastases. MAVERICK was designed to address that uncertainty with random assignment and scheduled imaging in both study groups.
Clinical references have continued to list MAVERICK as a key phase III comparison as practice evolves; the National Library of Medicine’s small-cell lung cancer reference describes the trial’s MRI-surveillance-versus-PCI framework. For now, the new results provide the strongest randomized evidence reported for avoiding PCI in a closely monitored, MRI-negative population, while leaving the final survival comparison to longer follow-up.