Abstract P1-18-07: Impact of PIK3CA mutation (PIK3CA-mt) clonality on alpelisib (ALP) activity based on real-world evidence (RWE) following liquid biopsy testing

Abstract

Abstract Background:ALP is an alpha-selective PI3K-inhibitor approved in combination with fulvestrant for PIK3CA-mt HR+/HER2- advanced breast cancer (aBC). These mutations may either be truncal (clonal) or acquired (subclonal) under treatment pressure; however, data regarding the efficacy of ALP in these two populations are currently limited. This study utilized RWE to assess how the PIK3CA genomic environment impacts ALP response. Methods: RWE was sourced from the GuardantINFORM (Guardant Health) database, which comprises aggregated commercial payer health claims and de-identified records from over 100,000 individuals with comprehensive ctDNA results via Guardant360 (G360). All HR+/HER2- aBC patients with one or more of the 11 PIK3CA-mt cited in the Therascreen PIK3CA RGQ PCR Kit ALP companion diagnostic approval (P190001) identified on a G360 since May 2019 were included. Patients must have had at least one claim of ALP after the index G360 test. Patients who received ALP claim(s) in the six months prior to their G360 test were excluded. PIK3CA-mt were defined by clonal fraction (copy number-adjusted PIK3CA mutation allelic fraction/maximum somatic mutation allelic fraction) >50% (clonal) or ≤50% (subclonal). Real-world time to discontinuation (rwTTD) and real-world time to next treatment (rwTTNT) were assessed as proxies for progression free survival. Log-rank tests were used to assess differences in rwTTD and rwTTNT and Chi-squared tests were used to compare the proportion of PIK3CA-mt and other co-occurring alterations between patients with only clonal and only subclonal PIK3CA-mt. Results:Of 223 eligible patients, 216 (96%) had no prior ALP exposure and were included for further analysis. Most patients had one PIK3CA-mt (199, 73%); 177 (82%) harbored only clonal mutations, 34 (16%) harbored only subclonal mutations, 5 (2%) harbored both. We saw no significant difference in rwTTD or rwTTNT for ALP in patients with clonal vs. subclonal PIK3CA-mt [median months to discontinuation = 5.0 (95% CI 4.0 - 6.9) vs. 7.4 (95% CI 3.7 - 11.1) p=0.82; median months to next treatment =7.0 (95% CI 5.5-9.4) vs. 9.0 (95% CI 4.0-12.6) p=0.81]. We observed no significant differences in the frequency of co-occurring alterations between samples with clonal vs. subclonal PIK3CA-mt (Table 1). Many alterations known to be associated with resistance to ALP and/or CDK4/6 inhibitors were identified, including RB1 and PTEN loss of function mutations. Patients with only subclonal PIK3CA-mt had a significantly higher proportion of E545K and E545G alterations compared to patients with only clonal PIK3CA-mt (E545K: 44% vs 26%, p=0.03; E545G: 6% vs 1%, p=0.017). Conclusions:Examination of RWE in patients treated with ALP after identification of PIK3CA-mt on G360 showed no significant difference in treatment outcomes or co-occurring mutations for clonal vs. subclonal PIK3CA-mt, suggesting that patients with PIK3CA-mt should be considered for ALP therapy irrespective of mutation clonality. While this study focused on outcomes related to PIK3CA hotspot alterations, a significant percentage of patients have PIK3CA non-hotspot alterations; assessment of ALP outcomes in this population is warranted. Table 1.Frequency of co-occurring alterations by PIK3CA-mt clonalityGeneClonal (N=177)Subclonal (N=34)p valueNo.%No.%TP538649%1441%0.428ESR18146%1338%0.419ATM3319%926%0.295EGFR3218%824%0.458RB12514%412%0.714PTEN2212%412%0.914FGFR12112%515%0.644FGFR22011%412%0.938MET1810%26%0.434SMAD4169%13%0.232ARID1A158%515%0.256APC148%39%0.858BRAF148%26%0.683GATA3137%26%0.761KRAS137%39%0.765BRCA1127%39%0.671KIT116%13%0.450CDK12106%13%0.515AR85%39%0.301BRCA285%26%0.732 Citation Format: Dejan Juric, Caroline Weipert, Leslie Bucheit, Rebecca Nagy, Justin Odegaard, Junhua Yu, Nicole Zhang, Jiemin Liao. Impact of PIK3CA mutation (PIK3CA-mt) clonality on alpelisib (ALP) activity based on real-world evidence (RWE) following liquid biopsy testing [abstract]. In: Proceedings of the 2021 San Antonio Breast Cancer Symposium; 2021 Dec 7-10; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2022;82(4 Suppl):Abstract nr P1-18-07.