Machine learning enables detection of early-stage colorectal cancer by whole-genome sequencing of plasma cell-free DNA.

BACKGROUND: Blood-based methods using cell-free DNA (cfDNA) are under development as an alternative to existing screening tests. However, early-stage detection of cancer using tumor-derived cfDNA has proven challenging because of the small proportion of cfDNA derived from tumor tissue in early-stage disease. A machine learning approach to discover signatures in cfDNA, potentially reflective of both tumor and non-tumor contributions, may represent a promising direction for the early detection of cancer. METHODS: Whole-genome sequencing was performed on cfDNA extracted from plasma samples (N = 546 colorectal cancer and 271 non-cancer controls). Reads aligning to protein-coding gene bodies were extracted, and read counts were normalized. cfDNA tumor fraction was estimated using IchorCNA. Machine learning models were trained using k-fold cross-validation and confounder-based cross-validations to assess generalization performance. RESULTS: In a colorectal cancer cohort heavily weighted towards early-stage cancer (80% stage I/II), we achieved a mean AUC of 0.92 (95% CI 0.91-0.93) with a mean sensitivity of 85% (95% CI 83-86%) at 85% specificity. Sensitivity generally increased with tumor stage and increasing tumor fraction. Stratification by age, sequencing batch, and institution demonstrated the impact of these confounders and provided a more accurate assessment of generalization performance. CONCLUSIONS: A machine learning approach using cfDNA achieved high sensitivity and specificity in a large, predominantly early-stage, colorectal cancer cohort. The possibility of systematic technical and institution-specific biases warrants similar confounder analyses in other studies. Prospective validation of this machine learning method and evaluation of a multi-analyte approach are underway.

An FDA Perspective on the Regulatory Implications of Complex Signatures to Predict Response to Targeted Therapies.

As technologies evolve, and diagnostics move from detection of single biomarkers toward complex signatures, an increase in the clinical use and regulatory submission of complex signatures is anticipated. However, to date, no complex signatures have been approved as companion diagnostics. In this article, we will describe the potential benefit of complex signatures and their unique regulatory challenges, including analytic performance validation, complex signature simulation, and clinical performance evaluation. We also will review the potential regulatory pathways for clearance, approval, or acceptance of complex signatures by the FDA. These regulatory pathways include regulations applicable to in vitro diagnostic devices, including companion diagnostic devices, the potential for labeling as a complementary diagnostic, and the biomarker qualification program. Clin Cancer Res; 23(6); 1368-72. ©2016 AACR.

FDA Approval Summary: Olaparib Monotherapy in Patients with Deleterious Germline BRCA-Mutated Advanced Ovarian Cancer Treated with Three or More Lines of Chemotherapy.

On December 19, 2014, the FDA approved olaparib capsules (Lynparza; AstraZeneca) for the treatment of patients with deleterious or suspected deleterious germline BRCA-mutated (gBRCAm) advanced ovarian cancer who have been treated with three or more prior lines of chemotherapy. The BRACAnalysis CDx (Myriad Genetic Laboratories, Inc.) was approved concurrently. An international multicenter, single-arm trial enrolled 137 patients with measurable gBRCAm-associated ovarian cancer treated with three or more prior lines of chemotherapy. Patients received olaparib at a dose of 400 mg by mouth twice daily until disease progression or unacceptable toxicity. The objective response rate (ORR) was 34% with median response duration of 7.9 months in this cohort. The most common adverse reactions (≥20%) in patients treated with olaparib were anemia, nausea, fatigue (including asthenia), vomiting, diarrhea, dysgeusia, dyspepsia, headache, decreased appetite, nasopharyngitis/pharyngitis/upper respiratory infection, cough, arthralgia/musculoskeletal pain, myalgia, back pain, dermatitis/rash, and abdominal pain/discomfort. Myelodysplatic syndrome and/or acute myeloid leukemia occurred in 2% of the patients enrolled on this trial.