Personalized medicine for head and neck cancer

Science Spotlight

Personalized medicine for head and neck cancer

from the Mendez lab, Clinical Research Division, and Kemp lab, Human Biology Division

June 18, 2018

Head and neck squamous cell carcinoma (HNSCC) arise from squamous cells in the outer skin layer or mucous membranes in the head and neck area. According to the National Institutes of Health, 600 000 new cases develop each year including 50 000 in the United States. One out of every two cases will be fatal. Genetic mutations play a critical role in cancer development. Mutations in the tumor suppressor TP53 are detected in 74% of HNSCC patients and are associated with poor clinical outcomes. However, many other mutations exist in the genome of HNSCC patients, albeit at a lower frequency, that could also contribute to treatment resistance. Characterization of these mutations and their drug sensitivity is essential to identify the optimal treatment for each individual.

To investigate the genomic landscape in HNSCC patients, a collaboration between the late Dr. Eduardo Mendez (Clinical Research Division), and Dr. Christopher Kemp (Human Biology Division) and staff scientist Dr. Chang Xu, performed in depth analyses of the mutational profile and tumor sensitivity of an aggressive case of treatment-resistant and p53-mutated HNSCC patient. The results of this study were recently published in the journal Clinical Cancer Research. The strength of the procedure is the ability to perform functional genomic analyses and drug testing on cells from the patient that will receive personalized medicine treatment. In addition to mutations characteristic of a p53-mutated HNSCC tumor, new targetable mutations critical for tumor development (driver mutations) were identified and discriminated from other bystander (non-driver) mutations. Such investigations are crucial to identify treatments that are most likely to succeed or not in individual patients.

Personalized medicine is possible when tumor biopsies are available from the patient. From these cells, mutated genes of interest that might be sensitive to therapeutic drugs are identified by genomic characterization of the mutational landscape (sequencing), comparison with other normal and tumor cells and siRNA screening. The optimal treatment is then identified by functional drug testing, before clinical applications.

Personalized medicine is possible when tumor biopsies are available from the patient. From these cells, mutated genes of interest that might be sensitive to therapeutic drugs are identified by genomic characterization of the mutational landscape (sequencing), comparison with other normal and tumor cells and siRNA screening. The optimal treatment is then identified by functional drug testing, before clinical applications.

Figure provided by Dr. Christopher Kemp.

Cell cultures were established from a biopsy sample and genomic analyses using whole exome sequencing (analysis of mutation spectra, copy number aberrations and gene expression), RNA sequencing, comparative genome hybridization, high-throughput siRNA phenotyping were performed. In addition, samples were subjected to the Clinical Laboratory Improvement Amendments (CLIA) certified PARIS test, a clinically compatible procedure that tests sensitivity to over 100 oncology relevant drugs. Identified candidate target genes with a relevant mutation profile and with known available drugs were selected and tested as candidates of interest by siRNA targeting and functional drug assessment. The identified genes were also validated in p53-mutated HSNCC cell lines and compared to non-tumorigenic keratinocyte cell cultures.

Numerous mutations (581) were identified in this patient including 102 indels (insertions or deletions) in 210 genes. Copy number amplifications and deletions were identified in 1299 and 831 genes, respectively. RNAseq detected 773 genes with increased expression and 12063 genes with decreased expression. These observations were compared to normal non-tumorigenic cells lines and data from the TCGA database for HNSCC tumor samples. Among the eighteen genes of interest identified, candidates such as the growth factor FGF3, the tumor suppressor CDKN2a or oncogenes PIK3CA and CCND1 were also frequently mutated in the TCGA database HNSCC cohort, suggesting their role as driver mutations.

To prioritize the candidate genes that would represent appropriate targets for treatment, further validations were performed on the same cells by siRNA targeting and functional drug testing. A siRNA library targeting 6659 genes was used. Targeting of 391 of these genes affected tumor cell viability. Knockdown of 5% of copy number amplified genes and 7.6% of genes with increased expression affected tumor cell viability suggesting that most identified mutations might not have phenotypic consequences. A second siRNA screen found that 23 out of the 174 tested genes affected specifically HNSCC tumor cell viability but not non-tumorigenic cell lines, with functions ranging from ubiquitin proteasome (protein recycling machinery) to cell cycle regulators, antiapoptotic MCL1 pathway or Wnt pathway.

To validate anticipated treatment response, experimental drug profiling was performed using 41 small molecules, including drugs that were FDA approved, in clinical development or experimental. Several drugs such as paclitaxel, vincristine (chemotherapies) or AZD1775 (targeted therapy against WEE1 cell cycle regulator in clinical trial and developed by Dr. Mendez) presented increased efficiency on tumor cells relative to non-tumorigenic cell line. In agreement with the siRNA data, the drugs presenting the highest efficiencies targeted genes involved in functions such as G2/M regulation, mitotic spindle and proteasome function.

Dr. Xu explained that “Establishing patient derived primary cell culture will allow us to perform functional assays in a few weeks”. Such experiments represent a considerable asset to increase positive outcomes not only in HNSCC patients but also for many types of cancer for which live biopsies are a possibility. According to Dr. Kemp, “in the near future, live tumor sample biopsies will be routine for use in a variety of functional assay”. To help make this possible, he and other researchers have created the Society for Functional Precision Medicine, whose goal is to increase awareness of the utility of performing functional testing in a clinical setting.

 

Funding for this study was provided by the National Institutes of Health, the American Cancer Society the Howard Hughes Medical Institute Early Physician-Scientist Career Development Award.

 

Xu C, Nikolova O, Basom R, Mitchell RM, Shaw R, Moser R, Park H, Gurley KE, Kao M, Green CL, Schaub FX, Diaz RL, Swan HA, Jang IS, Guinney J, Gadi VK, Margolin AA, Grandori C, Kemp CJ, Mendez E. 2018. Functional precision medicine identifies novel druggable targets and therapeutic options in head and neck cancer. Clinical Cancer Research. [Epub ahead of print].