Sep 25 2024 7 mins
In this JCO Precision Oncology Article Insights episode, Mitchell Elliot summarizes an editorial: “A Targeted Methylation–Based Multicancer Early Detection Blood Test Preferentially Detects High-Grade Prostate Cancer While Minimizing Overdiagnosis of Indolent Disease” by Dr. Brandon A. Mahal, et al. published on August 29, 2024.
TRANSCRIPT
Mitchell Elliott: Hello and welcome to JCO Precision Oncology Article Insights. My name is Mitchell Elliott, a JCO Editorial Fellow. Today, I'll be discussing the article, “A Targeted Methylation–Based Multicancer Early Detection Blood Test Preferentially Detects High-Grade Prostate Cancer While Minimizing Overdiagnosis of Indolent Disease,” by Mahal et al.
Cancer overdiagnosis, particularly of low-risk conditions that are unlikely to cause harm, is a common issue in screening tests. In prostate cancer screening, overdiagnosis affects 23% to 42% of cases, often due to the prevalence of low-grade cancers and the low specificity of the prostate specific antigen or PSA tests. Data from previous studies have highlighted that men with low grade prostate cancer often die with prostate cancer and not of prostate cancer. Over diagnosis can lead to unnecessary treatments, increased patient anxiety, side effects, and excessive healthcare costs. Multicancer early detection, or MCED tests offer a new approach by detecting multiple cancer types from a single blood sample with low false positive rates, typically less than 2%, and they also have the ability to predict the cancer type from this one test. The GRAIL Galleri test, based on methylation patterns of circulating tumor DNA, showed high accuracy detecting over 50 cancer types, including prostate cancer, in the circulating cell free genome atlas or CCGA in PATHFINDER studies. This type of MCED test paradigm is being developed for use alongside traditional screening methods in adults over the age of 50. This study evaluated this particular MCED test ability to detect both indolent and aggressive prostate cancer, aiming to assess its potential to contribute to over diagnosis.
This cohort was part of the circulating cell free genome atlas or CCGA study, a multicenter case control study with three phases to validate this particular MCED test. The CCGA enrolled 15,254 participants, of which 8,584 had cancer and 6,670 did not. Enrollment was carried out in 142 North American sites between 2016 and 2019. Eligibility for cancer cases required a confirmed diagnosis or high suspicion with planned biopsy or surgery within six weeks of enrollment. This study evaluated 420 recently diagnosed men with prostate cancer from substudy 3, the independent clinical validation arm. The PATHFINDER study was a prospective cohort study of 6,662 adults over the age of 50 enrolled from seven US health networks between December 2019 and December 2020. Participants underwent testing with the GRAIL Galleri test, with results shared with physicians and participants. The test indicated the presence or absence of a cancer signal and predicted the cancer signal of origin if detected. This study's prostate cancer cohort included 18 men diagnosed through MCED testing or PSA screening, excluding two with recurrent disease. PSA testing was not collected in this particular study.
Detectability by the Gleason group, clinical stage, association of detection status with tumor methylation fraction, and overall survival were assessed in these studies. The results are broken down by each substudy evaluated. Substudy three of the CCGA enrolled a clinically relevant patient population. The median age of the men enrolled were 65. Ethnic diversity was not represented, however, in this cohort, with only 15% of participants reporting as non-white, non-Hispanic. It is important to note that only 8.4% of patients included in the study self-identified as black non-Hispanic, a particular group of participants with a higher incidence in more aggressive prostate cancer.
The overall MCED test sensitivity for prostate cancer detection was low in 11.2% or 47 out of 420 patients included in this cohort. The cancer signal of origin prediction accuracy was 91.5% with 43 of 47 patients having prostate cancer predicted. The test did not detect any low-grade tumors. It detected 3 of 157 favorable or intermediate grade tumors as well as 4 of 78 unfavorable intermediate grade tumors, and finally 36 of 113 high grade tumors, typically, Gleason score 4 and 5. Detection increase was staged with only 3.2% or 3 of 95 of stage one disease detected with the MCED test, while 14.9% or 7 of 47 with stage 3 and 81.5% 22 out of 27 patients with stage four disease. Compared with expected overall survival estimated from the United States SEER database, non-detected cancer cases had roughly three times better overall survival with a hazard ratio of 0.263 with a 95% interval of 0.1 to 0.5 with a p value of less than 0.05, and detected case that had similar survival, the hazard ratio of 0.672 with a 95% interval crossing one and a p value of 0.2 when adjusted for age, Gleason grade, and clinical stage. This suggests that patients identified to be ctDNA positive at diagnosis have an overall worse outcome than those who are ctDNA negative, a consistent phenomenon with previous studies using the same or different tumor informed and diagnostic ctDNA assays.
Next, the authors evaluated the outcomes in the PATHFINDER cohort of 18 participants. The characteristics of patients enrolled were similar to the previous cohort. Only one case was detected, which was between Gleason group 3 and 5, and had either stage 3 or stage 4 disease not defined in the manuscript. Because only a single case of prostate cancer was identified in PATHFINDER via this test, cancer signal of origin, predicted accuracy, tumor methylation fraction, and survival outcomes were not calculated. In summary, this test preferentially detected high grade and advanced stage prostate cancers, identifying 93% of Gleason grade 3 to 5 and 67% of stages 3 and 4 cases, while notably did not detect Gleason grade 1, having only 1.9% of Gleason grade 2 detected and 4.2% of stage 1 and stage 2 cancers overall. Importantly, around one third of the detected cases in substudy three of the CCGA, involved non metastatic disease, including stage 1 to stage 3 were Gleason grade 3 to 5, which are potentially curable. Prostate cancers that were not detected via this test had better survival rates after adjusting for age, grade and stage in the SEER database. This suggested that MCED testing is unlikely to contribute to the overdiagnosis of indolent prostate cancers. Additionally, a positive cancer signal with a predicted prostate origin strongly indicates the presence of aggressive disease, warranting immediate diagnostic investigation.
One limitation of the study is the lack of representative inclusion of patients from diverse ethnic backgrounds. Overdiagnosis of prostate cancer due to PSA levels disproportionately affects black men, and the generalizability of these findings in the study is limited by the fact that over 85% of the study cohort was self-reported as white non-Hispanic. Further data is required to understand the biology of cancer in this community and limit the bias of molecular screening tests so they are effective regardless of ethnicity.
Thank you for listening to JCO Precision Oncology Article Insights. This was a summary on “A Targeted Methylation–Based Multicancer Early Detection Blood Test Preferentially Detects High-Grade Prostate Cancer While Minimizing Overdiagnosis of Indolent Disease.” Please follow and subscribe on your favorite streaming platforms. For more podcasts from ASCO, visit www.asco.org/podcasts.
The purpose of this podcast is to educate and to inform. This is not a substitute for professional medical care and is not intended for use in the diagnosis or treatment of individual conditions.
Guests on this podcast express their own opinions, experience, and conclusions. Guest statements on the podcast do not express the opinions of ASCO. The mention of any product, service, organization, activity, or therapy should not be construed as an ASCO endorsement.