Jacob J. Adashek, DO of Moffit Cancer Center gives an overview from ASCO 2020 the abstract entitled Targetable immune checkpoint molecules may be significantly differentially expressed in minority ethnicities.
Effects of immune control point blockade therapy (ICT) trials were primarily conducted in patients with melanoma and lung cancer, both of which are enriched for White patients. A National Cancer Database report, for example, showed that 97 per cent of melanoma first-line ICT therapies were given to White recipients (Patel, ASCO-SITC 2020). Because of the expanding indication of tumor types affecting more diverse populations, we tried to research whether minority populations might be expected to have different blockade response rates at the checkpoint.
TCGA received race and RNAseq expression profiles and primary site information for 7087 patients. Ethnicity has been tested for association with RNA expression of targetable control point genes (PD1, PDL1, PDL2, CTLA4, IDO1, LAG3, TIM3, TIGIT, OX40, VISTA, and GITR) in 5 forms of Wilcoxon tumor histology with Benjamini-Hochberg multiple-hypothesis correction. A dataset of > 2700 cases was collected from NantHealth, with exome / RNAseq data paired in whole. Ethnicity was allocated to 579 patients using allele-fraction from ~250 polymorphic single nucleotide sites identified exclusively in 6 populations within the 1000 Genomes project. Within this dataset, associations of ethnicity / checkpoint found in TCGA were checked.
Ethnicity was not a factor in varying expression of checkpoint molecules in lung cancer within the TCGA cohort. Within melanomas, the expression of PDL1, CTLA4 and IDO1 in Asian patients was lower than in White patients (each p = 0.04). For multiple hypothesis testing these correlations did not remain relevant after correction. In Black patients, breast cancers had substantially higher expression of PD1, CTLA4, IDO1, LAG3, GITR, and OX40 relative to White patients, all remaining significant after correction (adj. p 3.7e-5 to 6.4e-3); Colon cancers in white patients displayed greater expression of PDL1/2, IDO1, LAG3, TIM3, and GITR (p 0.04 to 0.0017). In White patients IDO1 was slightly higher even after correction (adj. p = 0.03), and lower in Black patients (adj. p = 0.03).
Ethnicity can be a major factor in blockade therapies at effectiveness checkpoints. Patients with white breast cancer can be predicted to display reduced sensitivity to PD1 / CTLA4 blockade, whereas patients with black colon cancer can demonstrate reduced sensitivity to IDO1 therapies such as epacadostat. A biomarker-driven approach to patient selection can improve the ICT outcomes of the ethnic disparities.