265P - Tumor-agonistic genomic profiling

Background

Genomic profiling using next-generation sequencing becomes pervasive in clinical practice. However, there is no standardized interpretation about actionable gene alterations. Here, we focused on clinically important gene variants and evaluated the clinical impact.

Methods

We performed genomic profiling in 592 patients (tissue-based analysis in 525 and liquid-based analysis in 67 patients). Highly actionable gene alterations were defined as mutations, amplifications, and fusions that are classified as Level A evidence in Clinical Interpretations of Variants in Cancer (CIViC) and Category 1A in the MD Anderson Knowledge Base for Precision Oncology. We analyzed the proportion of highly actionable gene alterations and the drug accessibility across cancer types in which drugs are not approved by insurance.

Results

Nine gene variants were classified into highly actionable gene alterations. On tissue-based analysis, 85 cases had any highly actionable gene alteration. Fifty-four out of 85 cases (63%) were not approved by insurance, and the most common variants were BRAF V600E and ERBB2 amplification. The tumor types with the highest rate of highly actionable gene alterations were thyroid papillary carcinoma (100%). Twelve out of the 54 patients (22%) joined clinical trials and received drugs not approved by insurance. The major drugs were agents targeting HER2 amplification and BRAF inhibitors. The most common reason not accessing to clinical trials was decline of performance status (20%). On the other hand, only 3 cases had highly actionable gene alterations on liquid-based analysis (4%) and of 1 case (33%), ERBB2 amplification, was received a drug not approved by insurance.

Conclusions

This study detected BRAF V600E and ERBB2 amplification across various types of cancer in clinical practice, and showed that those patients potentiated the access to clinical trials. It may be important for physicians to know highly actionable gene alterations detected in each cancer type and perform genomic profiling actively in patients with promising types of cancers.

Clinical trial identification

Editorial acknowledgement

Legal entity responsible for the study

The authors.

Funding

Has not received any funding.

Disclosure

K. Takeuchi: Financial Interests, Personal, Sponsor/Funding: Fujirebio, Daiichi Sankyo; Financial Interests, Personal, Invited Speaker: Eli Lilly, Chugai, Kyowa Kirin. S. Takahashi: Financial Interests, Personal, Invited Speaker: ONO Pharmaceutical Co., Bristol Myers Squibb, Chugai, Bayer; Financial Interests, Personal, Expert Testimony: MSD, AstraZeneca. All other authors have declared no conflicts of interest.