Abstract Background: PI3K/AKT signaling pathway is often activated in breast cancer (BC), through mutations in PIK3CA or AKT1, and alterations in PTEN, thus promoting cell survival. PI3K and AKT inhibitors have been shown to have significant activity against tumor progression and to overcome resistance in BC. We aim to explore the prevalence of PI3K pathway alterations and co-expression with other markers in different BC subtypes. This approach could allow identification of novel drug combinations that have the potential to elicit synergistic growth inhibition and be further explored in the clinical trial setting. Methods: Molecular profiles of 4895 female and male BC cases submitted to Caris Life Sciences were reviewed to identify pathogenic or presumed pathogenic mutations in PIK3CA, AKT1, PTEN, PIK3R1 and PIK3R2 using 592-gene next generation sequencing (NGS; average read depth 500 ×). Cases with loss of PTEN by IHC were also included. Co-mutation frequency with PIK3CA-AKT1-PTEN alterations were examined, including homologous recombination deficiency (HRD) genes and DNA damage response (DDR) pathways, markers of response to immune-oncology (IO) agents and RAS signaling pathway. Cases were classified in BC subtypes according to the ASCO-CAP guidelines. Results: The median age was 58 (range: 17-90). Gender frequency and primary/ metastatic site frequency were similar across BC subtypes. 3558 (72.7%) cases had at least one alteration in the PIK3CA-AKT1-PTEN pathway: 1472 (30.1%) cases harbored a PIK3CA mutation, 174 (3.6%) harbored an AKT1 mutation, and 2682 (54.8%) had PTEN alterations (PTEN mutation in 344, 7.0% and/or PTEN loss by IHC in 2516, 51.4% of cases). The most common hotspot mutations in PIK3CA were in the kinase domain (H1047R in 567 cases, 38.5% of all PIK3CA alterations) and in the helical domain (E545K in 304 cases, 20.7% of all PIK3CA alterations). Uncommon activating PIK3CA mutations were seen in 8.4% of breast tumors. The single hotspot mutation, E17K, was the most common AKT1 mutation (n=164, 94.3% of all AKT1 mutations). 81 (1.7%) tumors harbored a PIK3R1 mutation, with 66 unique alterations identified (4 pathogenic, 55 presumed pathogenic), and 4 (0.08%) cases harbored a PIK3R2 mutation (G373R mutation). With respect to the different BC subtypes, PIK3CA was the most frequent alteration in HER2 positive BC (present in 96.2% of mutated cases). Within HER2 negative subtypes, PTEN was most frequently altered and PTEN mutation or PTEN loss by IHC was present in 79.3% of mutated cases. Triple negative breast cancer (TNBC) was the subtype with the lowest frequency of PIK3CA mutations (18.0% in TNBC vs. 37% in other subtypes). The frequency of selected co-mutations with PIK3CA-AKT1-PTEN alterations is illustrated in table 1. Notable co-alterations in this cohort include increased PD-L1 expression and high tumor mutational burden (TMB). There were no significant increases in the frequency of mutations in DDR pathway mutations (not shown) in the PIK3CA-AKT1-PTEN altered cohort. Conclusions: We observed a high prevalence of expected hotspot mutations in PIK3CA and AKT1 across BC subtypes. There was a significant increase in PD-L1 expression in tumor cells, and high TMB in PIK3CA-AKT1-PTEN mutated cohorts. A similar association was seen with RAS signaling pathways. Further development of drugs that affect the PIK3CA-AKT1-PTEN pathway in all BC subtypes and combination with drugs that target the immune system may be of interest. Table 1. Selected co-alterations based on statistical significancePathwayGene/ ProteinAll subtypes (%)HR+ HER2+ (%)HR- HER2+ (%)HR+HER2- (%)TNBC (%)MTWTMTWTMTWTMTWTMTWTHomologous recombinationBRCA13.02.71.00.60.00.80.8*1.8*6.16.0BRCA24.15.41.02.53.62.34.1*7.9*4.43.0PALB20.7*1.3*0.00.00.00.00.5*2.5*0.90.3Possible predictors of IO benefitPD-149.850.038.552.266.766.737.541.766.058.1PD-L1 (SP142)6.9*4.2*3.30.78.14.93.33.011.68.1(Tumor Cells)PD-L1 (SP142)29.226.650.018.825.027.312.916.742.140.4(Immune Cells)MSI0.70.60.00.01.20.80.80.40.71.2TMB-High (≥ 10 mut/Mb)22.9*18.8*36.0*17.2*29.825.921.2*16.6*23.621.3Chromatin remodelingARID1A12.4*18.6*19.423.87.118.917.121.55.89.6ARID20.70.80.01.90.00.70.80.60.60.9RAS-RAF-MEK-ERK HRAS0.7*0.1*0.00.00.00.00.20.11.40.3KRAS2.0*1.1*1.00.00.00.01.91.22.22.0NRAS0.20.00.00.00.00.00.20.00.30.0BRAF0.50.11.00.01.20.00.60.30.40.0OthersTP5360.9*53.1*66.360.885.586.738.4*25.6*84.786.4CDH110.3*6.1*9.1*2.5*4.82.215.1*9.6*4.9*2.3*NF16.2*2.1*9.5*0.8*2.73.65.8*1.8*6.7*2.8*RB15.5*2.6*2.21.33.63.03.9*1.7*7.85.0ERBB22.3*3.4*3.03.13.56.73.23.51.0*2.3*Notes: MT = at least 1 pathogenic mutation in PIK3CA, AKT1, or PTEN or PTEN loss by IHC. WT = no pathogenic mutations in PIK3CA, AKT1, and PTEN.*statistically significant difference between MT and WT Citation Format: Katia Khoury, Antoinette Tan, Andrew Elliott, Joanne Xiu, Zoran Gatalica, Arielle L Heeke, Claudine Isaacs, Paula R Pohlmann, Lee S Schwartzberg, Michael Simon, Michael W Korn, Sandra M Swain, Filipa Lynce. Prevalence of phosphatidylinositol-3-kinase (PI3K) pathway alterations and co-alteration of other markers in breast cancer [abstract]. In: Proceedings of the 2019 San Antonio Breast Cancer Symposium; 2019 Dec 10-14; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2020;80(4 Suppl):Abstract nr P4-09-04.
Read full abstract