The NHS-Galleri trial, a landmark, randomized, controlled trial in multi-cancer early detection (MCED) research, recently reported its primary endpoint. The study, which enrolled over 142,000 asymptomatic participants aged 50–77 across England, added annual Galleri® (GRAIL, Inc) MCED cell-free DNA testing of peripheral blood samples to standard UK National Health System (NHS) cancer screenings over 3 years. The study’s hypothesis was that increased testing in an aging general population would lead to a reduction in late-stage (ie, stage III or higher) cancer diagnoses. While NHS-Galleri failed to meet its primary endpoint of producing a statistically significant reduction in combined stage III–IV cancer diagnoses, several of its secondary findings could be clinically meaningful, though full publication of the study data will be needed before definitive conclusions on the utility of routine MCED testing can be made.
An especially important secondary endpoint was achieved: a substantial reduction in stage IV cancer diagnoses with Galleri screening, with decreases exceeding 20% in the second- and third-year screenings for a prespecified group of 12 especially aggressive cancers (anus, bladder, colorectal, esophagus, head and neck, liver/bile duct, lung, lymphoma, myeloma/plasma cell neoplasm, ovary, pancreas, and stomach). The molecular testing also significantly decreased the rate at which cancers were first detected in an emergency setting. The reduced stage IV diagnoses occurred alongside a potentially clinically meaningful increase in stage I–II detection for these cancers. This increased rate of early detection could enable more-effective and better-tolerated therapeutic approaches for these especially aggressive cancers. In terms of sensitivity, Galleri produced a 4-fold higher overall cancer detection rate compared with standard screening alone for common cancers (breast, colorectal, cervical, and high-risk lung cancer).
While the underlying study hypothesis for NHS-Galleri lumped together stage III and IV cancers of all types, survival outcomes are considerably improved for patients diagnosed with stage III vs stage IV disease for many solid tumors. The increase in stage III diagnoses prevented the trial from achieving statistical significance for its primary endpoint. However, there is a growing therapeutic armamentarium for stage III cancers, some of which have curative potential, whereas stage IV (distant/metastatic disease) is still viewed as mostly incurable for the vast majority of cancers. Therefore, a shift toward stage III rather than stage IV at initial cancer diagnosis can be clinically meaningful, especially if it results in diagnosis while tumors can still be resected. Indeed, in the context of increasing use of neoadjuvant and adjuvant therapy approaches for some cancers, long-term disease control rates are improving even for higher-risk resections.
One clear outcome of the study was validation of the Galleri test’s performance metrics in a large general population, including positive predictive value and cancer signal origin accuracy. These were reported to have remained consistent with prior studies. Further insight into the potential benefit of this pan-tumor identification capacity vs the mostly presumptive tests for common malignancies is needed.
High Level
Many unanswered questions remain regarding the NHS-Galleri trial. Additional data on the clinical significance of detected cancers, potential overdiagnosis, and cost-effectiveness in real-world practice have yet to be released. The principle of stage IV diagnoses shifting toward earlier disease is promising, but definitive mortality benefit will require extended follow-up. Another important question, especially in US settings, is how positive results when cancers are undetectable in follow-up will impact the oncology community. All of these insights will be necessary to adequately assess whether implementation of widespread screening is prudent and which methodologies will be most feasible for this task.
Ground Level
The utility of molecular testing in oncology is rapidly expanding, driven by increases in sensitivity and decreases in cost and turnaround time. Molecular biomarker analyses now drive therapy optimization in multiple settings, and it is possible that improved molecular screening technologies could allow for earlier recognition of cancers, when they can be more effectively treated. Further research is needed to fully develop multiple emerging technologies, and there may be opportunities for collaborative research at the community level.
