Can New Algorithm Shed Light on Treatment Resistance in Head and Neck Cancer?

July 30, 2018

The CoGAPS gene activity algorithm may elucidate molecular changes that occur as H&N tumors acquire resistance to cetuximab, says a Johns Hopkins team.

Investigators of a multicenter in vitro study in Genome Medicinereport an approach aimed at improving therapy for head and neck squamous cell carcinoma (HNSCC), by tracking the evolution of treatment resistance. They examined how cancers acquire resistance to cetuximab over time and evaluated whether those changes could be modeled computationally to determine patient-specific resistance timelines.

The researchers assert that the Coordinate Gene Activity in Pattern Sets algorithm (CoGAPS) may help to determine key molecular changes that occur in conjunction with development of resistance to cetuximab. Senior study author Elana Fertig, PhD, said this computational model may result in better treatment of head and neck cancer, and ultimately better outcomes.

Fertig et al examined cetuximab treatment effects on head and neck squamous cell carcinoma (HNSCC) cell lines over a period of 11 weeks. CoGAPS was used to quantify the evolving changes during treatment. With the CoGAPS algorithm combining experimental biology and computer programming, the scientists hope to give clinicians and patients better information about how the disease is responding to therapy.

As the authors explained, “[W]e used an in vitro HNSCC cell line model to induce resistance and measure the molecular changes using high-throughput assays while the resistant phenotype developed. Gene expression and DNA methylation changes were screened weekly while acquired cetuximab resistance was induced in SCC25 cell line (intrinsic sensitive to cetuximab) and compared to the status of the untreated controls at the same culturing time point. CoGAPS inferred specific patterns of expression and DNA methylation that are associated with the gradual establishment of acquired cetuximab resistance.”

Fertig and colleagues found that transcriptional changes resulted from immediate therapeutic response or resistance. However, the study showed that epigenetic alterations only occurred with resistance. The authors concluded that “integrated analysis demonstrates delayed onset of changes in DNA methylation relative to transcription, suggesting that resistance is stabilized epigenetically.”

The Johns Hopkins researchers theorize that the current computational approach may be particularly useful in studying how cancer cells change over time during treatment with targeted therapies. 

Yang Xie, MD, PhD, Director of the Quantitative Biomedical Research Center at UT Southwestern and with the University’s Bioinformatics Core Facility in Dallas, Texas, said this approach could help to better guide the use of targeted therapies. Subsequently, it could enable selected patients to avoid treatments that may offer them no benefit. “It is very important but also challenging to study the time course of drug resistance, and this study attempts to address this question. The algorithm and the results published in this study are promising. If we can predict when the resistance will happen, physicians can start intervention earlier to prevent the drug resistance or change the treatment plan. This can improve patient outcomes,” Xie told Cancer Network.