Background: Molecular aberrations in the phosphatidylinositol 3-kinase (PI3K) pathway have been documented across cancers, especially PIK3CAmutations and mutation or loss of PTEN. These alterations may be relevant to therapies targeting the PI3K/PTEN/Akt/mTOR signaling pathway. Methods: Molecular profiling was performed on 19,784 tumors ( > 40 cancer types) at a CLIA-certified laboratory. Tests included next generation sequencing (NGS), protein expression (immunohistochemistry), and gene amplification (FISH or CISH). Results: Frequency and type of PIK3CA, AKT1 and PTEN mutations were collated across cancers. Aggregate gene mutation rates (47 genes), protein expression rates (18 proteins), and copy number (5 biomarkers) were measured. Comparison of frequencies and correlations across cancers identified lineage-specific differences, and co-incidences of associated biomarkers, which will be described. Of note, endometrial, breast, cervical, anal squamous cell, and bladder cancers had the highest PIK3CA mutation rate (37%, n = 1600; 31%, n = 2282; 29%, n = 284; 28%, n = 67, 22%, n = 303, respectively). Patterns in AKT1 and PTEN mutation rates differed by cancer, as did PTEN loss - hepatocellular, 57%, prostate, 52%, and endometrial 50% loss. Co-mutation of PTEN and PIK3CA occurred in 1.5% of breast, 0% of prostate, and 12% of endometrial cancers. Of interest, PIK3CA mutations and PTEN loss co-occurred frequently, e.g. 31% of PIK3CAmutated patients also have a PTEN loss. PIK3CA mutations across cancers were distributed 43% in exon 9, 33% in exon 20, and 24% in other exons. Distribution of PIK3CA mutations by cancer type varied and occurred more frequently in the absence of HER2 protein expression or copy number increase (p= 0.0001) and more frequently in the presence of hormone receptor overexpression (androgen receptor (AR), progesterone receptor (PR), and estrogen receptor (ER)) (p= 0.0335). PTEN loss was seen in 27% of patients with and 30% without HER2 overexpression or amplification (p= 0.004). Conclusions: Patterns of biomarker co-alterations across cancers may provide new insights relevant to targeted therapy and may be crucial to optimizing combination treatments.