Rapid responses are electronic comments to the editor. They enable our users
to debate issues raised in articles published on bmj.com. A rapid response
is first posted online. If you need the URL (web address) of an individual
response, simply click on the response headline and copy the URL from the
browser window. A proportion of responses will, after editing, be published
online and in the print journal as letters, which are indexed in PubMed.
Rapid responses are not indexed in PubMed and they are not journal articles.
The BMJ reserves the right to remove responses which are being
wilfully misrepresented as published articles or when it is brought to our
attention that a response spreads misinformation.
From March 2022, the word limit for rapid responses will be 600 words not
including references and author details. We will no longer post responses
that exceed this limit.
The word limit for letters selected from posted responses remains 300 words.
We write in response to some of the key points raised in the 4th January Barclay/Bradley editorial regarding the GRAIL-NHSE partnership.
GRAIL’s multi-cancer early detection (MCED) test represents one of the first examples of an innovation developed by a private company but intended for public health uptake. The resultant shortened timeline will be new to those of us who have studied and implemented cancer screening in the past. The benefit should be that innovations that work are made available to patients faster than would otherwise be possible.
Adding MCED to ongoing single cancer screening could dramatically improve early detection at the population level and thereby bend the cancer mortality curve. The collaboration with the NHS enables a large trial in the real-world setting. The NHS press release did not include detail regarding design; it will be a pragmatic, controlled study, with randomisation at the individual level.
We are designing the study for the primary endpoint of late-stage cancer incidence - a feasible endpoint to observe on this timeline and in this setting. We hope it will demonstrate the opportunity to significantly reduce the incidence of late-stage cancer. While mortality has historically been the primary endpoint of previous trials, we believe late-stage incidence at diagnosis to be an acceptable surrogate for mortality in the multi-cancer context. NHS England has also recognised this in its long-term plan, with an ambition that 75% of all stageable cancers should be detected at stage I or II by 2028. The design builds upon prior screening studies for lung, breast, and colorectal cancers that have linked detection at earlier stage to reductions in mortality, and leverages the improved recent understanding and recording of stage. We will model cancer-specific mortality based on the results of the study and will follow participants to assess impact on mortality, but mortality trend will be a secondary endpoint.
The editorial suggested that GRAIL’s CCGA case/control study may be insufficient for real-world populations with lower cancer prevalence. GRAIL has a progressive clinical development strategy. The CCGA study of over 15,000 participants enabled discovery and initial clinical validation. Large studies of intended use populations have followed, including a prospective cohort of almost 100,000 women (STRIVE) and an interventional study (PATHFINDER) of over 6,200 patients which completed enrolment last month. This strategy assures us that our validation will translate effectively into a real-world screening programme and has resulted in GRAIL having the largest known database of human methylation in participants with and without cancer. In CCGA, cancer cases across types and stages were matched carefully to a control group without cancer, and the study was enriched for confounding indications. This provided a large, diverse dataset to train the classifier and address potential challenges of overfitting, as often seen in small studies that don’t ultimately reproduce findings in larger populations. Based on the size, quality and robust methods of CCGA, our belief is that the validation results will be reproducible.
The editorial noted that evaluating a programme that screens people for many cancers simultaneously is difficult. We agree that evaluating MCED will be more complex than studies seeking to evaluate one cancer test at a time. With our world-class advisors and investigators, including several UK oncologists, primary care doctors, and other specialists, we are confident that this pragmatic randomised study design will be scientifically robust.
In conclusion, we believe that a properly-evaluated MCED test is our best patient-centric chance to make a significant reduction in cancer mortality, as we transition from only screening for individual cancers to screening individuals for many cancers.
Yours,
Sir Harpal Kumar
President, Grail Europe
Competing interests:
I am an executive and shareholder in Grail
Re: “Liquid biopsy” for cancer screening
Dear Editor
We write in response to some of the key points raised in the 4th January Barclay/Bradley editorial regarding the GRAIL-NHSE partnership.
GRAIL’s multi-cancer early detection (MCED) test represents one of the first examples of an innovation developed by a private company but intended for public health uptake. The resultant shortened timeline will be new to those of us who have studied and implemented cancer screening in the past. The benefit should be that innovations that work are made available to patients faster than would otherwise be possible.
Adding MCED to ongoing single cancer screening could dramatically improve early detection at the population level and thereby bend the cancer mortality curve. The collaboration with the NHS enables a large trial in the real-world setting. The NHS press release did not include detail regarding design; it will be a pragmatic, controlled study, with randomisation at the individual level.
We are designing the study for the primary endpoint of late-stage cancer incidence - a feasible endpoint to observe on this timeline and in this setting. We hope it will demonstrate the opportunity to significantly reduce the incidence of late-stage cancer. While mortality has historically been the primary endpoint of previous trials, we believe late-stage incidence at diagnosis to be an acceptable surrogate for mortality in the multi-cancer context. NHS England has also recognised this in its long-term plan, with an ambition that 75% of all stageable cancers should be detected at stage I or II by 2028. The design builds upon prior screening studies for lung, breast, and colorectal cancers that have linked detection at earlier stage to reductions in mortality, and leverages the improved recent understanding and recording of stage. We will model cancer-specific mortality based on the results of the study and will follow participants to assess impact on mortality, but mortality trend will be a secondary endpoint.
The editorial suggested that GRAIL’s CCGA case/control study may be insufficient for real-world populations with lower cancer prevalence. GRAIL has a progressive clinical development strategy. The CCGA study of over 15,000 participants enabled discovery and initial clinical validation. Large studies of intended use populations have followed, including a prospective cohort of almost 100,000 women (STRIVE) and an interventional study (PATHFINDER) of over 6,200 patients which completed enrolment last month. This strategy assures us that our validation will translate effectively into a real-world screening programme and has resulted in GRAIL having the largest known database of human methylation in participants with and without cancer. In CCGA, cancer cases across types and stages were matched carefully to a control group without cancer, and the study was enriched for confounding indications. This provided a large, diverse dataset to train the classifier and address potential challenges of overfitting, as often seen in small studies that don’t ultimately reproduce findings in larger populations. Based on the size, quality and robust methods of CCGA, our belief is that the validation results will be reproducible.
The editorial noted that evaluating a programme that screens people for many cancers simultaneously is difficult. We agree that evaluating MCED will be more complex than studies seeking to evaluate one cancer test at a time. With our world-class advisors and investigators, including several UK oncologists, primary care doctors, and other specialists, we are confident that this pragmatic randomised study design will be scientifically robust.
In conclusion, we believe that a properly-evaluated MCED test is our best patient-centric chance to make a significant reduction in cancer mortality, as we transition from only screening for individual cancers to screening individuals for many cancers.
Yours,
Sir Harpal Kumar
President, Grail Europe
Competing interests: I am an executive and shareholder in Grail