Validation of a blood test for multi-cancer risk stratification in a lung cancer screening cohort
Fahrmann, J.; Irajizad, E.; Rudsari, H.; Vykoukal, J.; Toumazis, I.; Khoramisarvestani, S.; Ansari, S.; Yang, J.; Kettner, N.; Dennison, J. B.; Ostrin, E.; Hanash, S.
Show abstract
PurposeWe report a blinded validation study of a ten-protein marker blood test for assessing risk of developing or harboring nine common cancers in a prospective lung cancer screening cohort. Patients and MethodsThe foundation of the multi-cancer risk stratification test (MCaST), test is a 4-marker protein panel consisting of ProSFTPB, CEA, CA125, and CYFRA-21 which has been extensively validated for risk of lung cancer and which has been expanded to include six additional markers to encompass, in addition to lung cancer, prostate, colorectal, breast, ovarian, pancreatic, liver, esophageal, and stomach cancers. Blinded validation samples consisted of 1,235 plasmas collected prior to diagnosis from 171 cancer cases and 526 randomly selected non-case controls. Fixed individualized combination rules as well as cancer-specific risk thresholds predefined based on established clinical guidelines were applied. ResultsAt the subject level, the MCaST tested positive for imminent risk of lung cancer in 65 of 73 lung cancer cases, including 45 of 51 early-stage cases, with a median time of 12.7 months (interquartile range [IQR]: 1.7 - 27.8 months) between the first positive MCaST and diagnosis of lung cancer. The positive rate of MCaST for other cancers was 88.9% for prostate, 63.6% for invasive breast, 85.7% for CRC, 50% for ovarian, 50% for pancreatic, and 67% for esophagus plus stomach cancers with a median (IQR) time of 20.7 months (10.2 - 42.8 months) from the first positive MCaST to a clinical diagnosis across the cancer types. Overall accuracy for tissue-of-origin (TOO) signal was 94.6%, and as high as 98.1% for lung cancer. ConclusionMCaST has utility for assessing cancer risk for common and lethal solid cancers in this smoker population.
Matching journals
The top 5 journals account for 50% of the predicted probability mass.
Similar papers in this journal
- A new colorectal cancer risk prediction model incorporating family history, personal and environmental factors 94%
- A Quantitative Framework to Study Potential Benefits and Harms of Multi-cancer Early Detection Testing 92%
- Hereditary haemochromatosis beyond liver cancer: increased risk of prostate cancer during an 11-year follow-up 91%
Similar papers in this journal
- Asian Lung Cancer Absolute Risk Models for lung cancer mortality based on China Kadoorie Biobank 94%
- Stratifying Lung Adenocarcinoma Risk with Multi-ancestry Polygenic Risk Scores in East Asian Never-Smokers 92%
- Genetic analysis of lung cancer reveals novel susceptibility loci and germline impact on somatic mutation burden 91%
Similar papers in this journal
- Combination of hotspot mutations with methylation and fragmentomic profiles to enhance Multi-Cancer Early Detection 93%
- Associations of the 2018 World Cancer Research Fund/American Institute of Cancer Research (WCRF/AICR) Cancer Prevention Recommendations with Stages of Colorectal Carcinogenesis 91%
- COVID-19 Outcomes in Patients with Cancer: Findings from the University of California Health System Database 91%
Similar papers in this journal
- Methylation scores for smoking, alcohol consumption, and body mass index and risk of seven types of cancer 93%
- Cancer and the risk of COVID-19 diagnosis, hospitalisation, and death: a population-based multi-state cohort study including 4,618,377 adults in Catalonia, Spain 92%
- Associations of plasma omega-6 and omega-3 fatty acids with overall and 19 site-specific cancers: a population-based cohort study in UK Biobank 91%
"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.