In cancer, early detection and accurate clinical decision-making are critical for the outcome. Unfortunately, current methods of diagnosis and monitoring are not always sensitive enough to detect cancer until it often progresses to certain dismal outcome, and ambiguous results can lead to misadministration of therapies or lack of treatment altogether.
Liquid biopsy is transforming cancer diagnosis and treatment monitoring by enabling the sequencing and analysis of circulating tumor cell free DNA (cfDNA). However, the abundance of cfDNA in early detection and residual post-operative disease is low and vastly outnumbered by DNA from healthy cells. Therefore, most sequencing approaches focus on deep targeted sequencing, which can result in false-negative readouts. In addition, high technical signal-to-noise ratio often confounds the accuracy of results.
With a phase I ASPIRE grant, Dan Landau and his collaborators took a different approach and designed a strategy called MRDetect (minimal residual disease) that takes advantage of information in the entire genome to increase sensitivity. He and his team replaced sequencing depth with breadth by combining whole genome sequencing with orthogonal information, such as single nucleotide variants and copy number variations. Underlying the analysis of these data is an artificial intelligence and advanced analytics platform that conducts in silico analysis of tumor and blood samples from individual patients.
They leveraged their richly annotated patient cohorts and integrated genome-wide mutation catalogue signatures to teach their machine-learning algorithms to distinguish between cancer-altered sequences and sequencing errors. This artificial intelligence detection method delivers patient-specific results quickly (within a week) and informs better clinical decisions without the need for patient-specific assays.
The group applied this approach to samples from lung adenocarcinoma and melanoma cohorts to establish clinical application feasibility for residual disease detection after surgery. This project resulted in a Nature Medicine publication, in which Dr. Landau and colleagues reported an increased sensitivity of circulating tumor DNA down to tumor fractions as low as 10-5 with just 35X whole genome sequencing coverage; this is 100X more sensitive than comparable techniques. Because this approach integrates single nucleotide variants and copy number variations, it is also applicable for tumors with high mutation load or aneuploidy.
Following the successful outcome of this project, Dr. Landau was awarded a second ASPIRE grant to demonstrate the feasibility of MRDetect for colorectal cancer in collaboration with Claus Andersen at Aarhus University in Denmark.
Dr. Landau has now co-founded C2i Genomics to make this software platform available to physicians and the drug discovery industry. Using the DNA signature of the first tumor excised, together with analytics from blood draws during and after therapy, the C2i platform helps physicians make real-time decisions about ongoing or future treatment approaches. For pharmaceutical companies and organizations conducting clinical trials, the C2i platform guides trial enrollment, enables real-time monitoring of treatment response, and provides insights into treatment recommendations.
The Mark Foundation participated in C2i’s Series A round of financing in June 2020, which raised $12 million. In 2021, C2i secured $100 million in venture-backed funding, which included an additional Mark Foundation investment, to scale up commercialization and expand their reach.
Zviran A, Schulman RC, Shah M, Hill STK, Deochand S, Khamnei CC, Maloney D, Patel K, Liao W, Widman AJ, Wong P, Callahan MK, Ha G, Reed S, Rotem D, Frederick D, Sharova T, Miao B, Kim T, Gydush G, Rhoades J, Huang KY, Omans ND, Bolan PO, Lipsky AH, Ang C, Malbari M, Spinelli CF, Kazancioglu S, Runnels AM, Fennessey S, Stolte C, Gaiti F, Inghirami GG, Adalsteinsson V, Houck-Loomis B, Ishii J, Wolchok JD, Boland G, Robine N, Altorki NK, Landau DA. Genome-wide cell-free DNA mutational integration enables ultra-sensitive cancer monitoring. Nat Med. 2020.