Abstract
INTRODUCTION
In patients with EGFR or KRAS-mutant lung adenocarcinomas, the prognostic impact of a concurrent PIK3CA mutation remains unclear. Although preclinical data suggest that sensitivity to EGFR tyrosine kinase inhibition (TKI) is decreased in EGFR-mutant lung cancers also harboring a PIK3CA mutation, this interaction has not been explored clinically.
METHODS
Patients with lung adenocarcinomas harboring a PIK3CA mutation concurrent with a separate driver mutation were identified via mutational hotspot testing, multiplex sizing assays, and FISH. Overall survival (OS) and outcomes with EGFR TKI monotherapy (EGFR-mutant) were estimated using Kaplan-Meier methods and compared between double mutant (EGFR or KRAS-mutant, concurrent PIK3C-mutant) and single mutant patients (EGFR or KRAS-mutant, PI3KCA wild-type) using log-rank tests.
RESULTS
In EGFR and KRAS-mutant lung cancers, a concurrent PIK3CA mutation was associated with a decrease in median OS: 18 vs 33 months (EGFR double n=10 vs single n=43 mutant, p=0.006), and 9 vs 16 months (KRAS double n=16 vs single n=47 mutant, p=0.020). In EGFR-mutant lung cancers, a concurrent PIK3CA mutation did not impact benefit from EGFR TKI monotherapy. Single vs double mutant: objective response rate 83% (n=29) vs 62% (n=6, p=0.80), median time to progression 11 (n=29) vs 8 months (n=6, p=0.84), and median duration of TKI therapy 15 (n=32) vs 15 months (n=10, p=0.65).
CONCLUSION
A concurrent PIK3CA mutation is a poor prognostic factor in patients with advanced EGFR- or KRAS-mutant lung adenocarcinomas. There was no evidence that clinical benefit from EGFR TKI monotherapy is affected by a concurrent PIK3CA mutation in EGFR-mutant lung cancers.
INTRODUCTION
PIK3CA mutations are found in approximately 7% of patients with lung adenocarcinomas.1 These mutations commonly occur in hotspots in exons 9 and 20, corresponding to the helical and kinase domains of p110α, the catalytic subunit of the PI3K enzyme. Mutations in these locations can result in constitutive activation of PI3K activity, phosphorylation of AKT, and downstream activation of mTORC1, which are essential for cell survival and proliferation.2, 3
In contrast to the mutual exclusivity of many oncogenic driver mutations in lung cancers, the co-occurrence of PIK3CA mutations with mutations in other oncogenic driver genes in lung adenocarcinomas is well described.4-6 The Lung Cancer Mutation Consortium (LCMC) identified a PIK3CA mutation via comprehensive molecular profiling in almost half of all lung cancers where a genomic alteration in more than one oncogene was found.7 Previous series have described the characteristics of PIK3CA-mutant lung cancers as a subgroup, but the role that PIK3CA mutations play as modifiers of clinical outcomes in EGFR- and KRAS-mutant lung cancers has not previously been defined.
From a therapeutic standpoint, an important question is whether or not PIK3CA mutations affect response to EGFR tyrosine kinase inhibitor (TKI) therapy in EGFR-mutant lung cancers. In vitro EGFR TKI sensitivity in EGFR-mutant lung cancer cell lines has been correlated with downregulation of the PI3K pathway.8 The introduction of the activating PIK3CA E545K mutation into a gefitinib-sensitive EGFR-mutant lung cancer cell line (HCC827, EGFR exon 19 deletion) via retroviral infection, for example, results in increased resistance to gefitinib and protection from gefitinib-mediated apoptosis.9 These preclinical data suggest that activation of the PI3K pathway can blunt the response to TKI therapy, but clinical series have not previously been reported.
We set out to determine the prognostic impact of PIK3CA mutations in patients whose lung cancers harbor a mutation in a second driver, particularly EGFR and KRAS, as these are the most common oncogenes in lung cancer.4, 10 In addition, we compared EGFR TKI therapy outcomes in patients with concurrent EGFR-mutant and PIK3CA-mutant lung cancers to patients with EGFR-mutant and PIK3CA wild-type lung cancers to elucidate the impact of PIK3CA mutations on response rate, overall survival, time to progression, and duration of TKI therapy.
METHODS
Patient Selection
Patients with newly diagnosed lung adenocarcinomas who were evaluated at Memorial Sloan Kettering Cancer Center between January 2009 and July 2013 were retrospectively identified as part of an institutional review board-approved study. Pathologic confirmation of lung adenocarcinoma histology was performed by the Department of Pathology. Patients who received systemic therapy prior to molecular diagnostic testing were excluded.
Targeted molecular profiling of tumor tissue was performed via an institutional algorithm under the Lung Cancer Molecular Analysis Project (LC-MAP) in the following order: 1) multiplex sizing assays for insertions and deletions in EGFR and ERBB2, 2) mutational hotspot testing by a mass spectrometry-based nucleic acid assay on the Sequenom platform for 91 mutations in 8 genes (EGFR, HER2, KRAS, NRAS, BRAF, MAP2K1, PIK3CA and AKT1), and 3) break apart fluorescence in situ hybridization tests for recurrent gene rearrangements involving ALK.
We identified subjects whose tumors harbored a PIK3CA mutation concurrent with one of the following genomic alterations: mutations in EGFR, KRAS, ERBB2, NRAS, BRAF, MAP2K1, and AKT1, and rearrangements involving ALK. These tumors are hereafter referred to as double-mutant (double-Mt). Three control groups of single-mutant (single-Mt) patients were also identified: (1) single EGFR-Mt (EGFR-mutant, PIK3CA wild-type), (2) single KRAS-Mt (KRAS-mutant, PIK3CA wild-type), and (3) single PIK3CA-mutant (PIK3CA-mutant, EGFR and KRAS wild-type). A review of clinicopathologic and molecular features was conducted.
The current study includes 19 patients from our previously published series of PIK3CA-mutant lung cancers4 (12 double-Mt and 7 PIK3CA single-Mt) as this paper did not examine prognosis in EGFR- and KRAS-mutant lung cancer subsets or response to EGFR TKI therapy in EGFR-mutant lung cancers.
Response Evaluation
Outcomes with EGFR TKI monotherapy were determined for all patients with a sensitizing EGFR mutation, including patients with EGFR single-Mt and EGFR double-Mt (concurrent EGFR and PIK3CA mutations) lung cancers. Radiographic review of all available imaging studies at baseline and on serial follow-up on therapy was performed by a thoracic radiologist (A.P.).
Patients were evaluable for radiographic response if they had both baseline imaging and at least one repeat evaluation after continuous EGFR TKI monotherapy. Best response to EGFR TKI monotherapy was categorized as a complete response (CR), partial response (PR), stable disease (SD), or progression of disease (POD) based on the Response Evaluation Criteria In Solid Tumors (RECIST) version 1.1.
Statistics
Overall survival (OS) and outcomes with EGFR TKI monotherapy including time to progression (TTP) and duration of therapy were calculated using the Kaplan-Meier method. OS was calculated from the date of diagnosis of metastatic disease until death. Patients who did not die during the study time were censored at the date of last available follow-up. TTP was calculated from the date of initiation of EGFR TKI monotherapy until radiographic progression via RECIST v1.1 on therapy. Patients who did not progress were censored at date of last follow-up on EGFR TKI.
Duration of therapy was calculated from the date of initiation until the last documented visit date on EGFR TKI monotherapy. Patients were censored if we did not have further follow-up on them beyond this date. Time to best response was calculated from the date of initiation of EGFR TKI monotherapy until date of best response by RECIST v1.1. OS, TTP, time to best response, and duration of therapy were compared between molecularly-defined subgroups using the log-rank test. Time to best response was compared between single-Mt vs double-Mt patients using the Wilcoxon Rank Sum Test. Fisher’s exact test was used to analyze the association between single- or double-Mt tumors and best response. All statistical tests were two-sided, and the level of significance was 5% (p<0.05). Statistical analyses were performed using R (version 3.0.1; R Development Core Team) with the “survival” package.
RESULTS
Patients
We identified 37 patients with oncogene-driven lung adenocarcinomas that harbored a concurrent PIK3CA mutation (Table 1). The majority of patients were female (70%, n=26/37), with a history of current or former cigarette smoking (76%, n=29/37), and stage IIIB/IV disease at diagnosis (65%, n=24/37).
Table 1.
The clinical and molecular features of PIK3CA-mutant lung cancers harboring a concurrent driver oncogene.
| PIK3CA Double-Mutant Lung Cancers (N=37) | |
|---|---|
| Age* (years) | 68 (38-89) |
|
| |
| Gender | |
| Male | 11 (30%) |
| Female | 26 (70%) |
|
| |
| Smoking History | |
| Never | 9 (24%) |
| Former | 24 (65%) |
| Current | 4 (11%) |
| Pack Years Smoked* | 31 (0-120) |
|
| |
| Stage at Diagnosis | |
| I-IIIA | 13 (35%) |
| IIIB/IV | 24 (65%) |
|
| |
| PIK3CA Mutation | |
| Exon 9 | |
| E542K | 12 (32%) |
| E545K | 12 (32%) |
| E545G | 1 (3%) |
| E545D | 1 (3%) |
| Exon 20 | |
| H1047R | 6 (16%) |
| H1047L | 5 (14%) |
|
| |
| Concurrent Mutation | |
| KRAS | 22 (60%) |
| EGFR | 13 (35%) |
| BRAF | 1 (3%) |
| MAP2K1 | 1 (3%) |
Median (range)
PIK3CA mutations were predominantly found in the exon 9 helical domain (70%, n=26/37; including E542K, E545K, 3545G, and E545D), and the remainder in the exon 20 catalytic domain (30%, n=11/37; including H1047R and H1047L). Genomic alterations that were concurrently identified with PIK3CA mutations included KRAS mutations (n=22, 59%), EGFR mutations (n=13, 35%), a BRAF V600E mutation (n=1, 3%), and a MAP2K1 K57N mutation (n=1, 3%).
KRAS was the most common concurrently mutated gene in patients with PIK3CA exon 9 mutations (69%, n=18/26) while EGFR was the most common concurrently mutated gene in patients with PIK3CA exon 20 mutations, (55%, n=6/11). No major differences in mutation type, gender, and smoking history were noted between EGFR double-Mt and single-Mt, and KRAS double-Mt and single-Mt patients (p>0.05, Supplementary Table 1).
Prognostic impact of concurrent PIK3CA mutations
In EGFR-mutant lung adenocarcinomas, median OS was shorter in patients with metastatic double-Mt vs single-Mt disease: 18.0 months (95% CI 13.9-NA months, n=10) vs 33.3 months (95% CI 28.2-NA months, p=0.006, n=43) (Figure 1A). Similarly, in KRAS-mutant lung adenocarcinomas, median OS was shorter in patients with metastatic double-Mt vs single-Mt disease: 8.9 months (95% CI 7.8-NA months, n=16) vs 16.2 months (95% CI 10.7-27.5 months, p=0.020, n=47) (Figure 1B).
Figure 1. Impact of PIK3CA Mutations on Overall Survival.
Overall survival (OS) of patients with EGFR single vs double-mutant (1A), KRAS single vs double-mutant (1B), EGFR double-mutant vs PIK3CA single-mutant (1C), and KRAS double-mutant vs PIK3CA single mutant lung cancers (1D).
Conversely, in PIK3CA-mutant lung adenocarcinomas, the presence of a concurrent EGFR mutation or KRAS mutation did not significantly affect median OS compared to patients without a concurrent mutation. Median OS was 18.0 months (95% CI 13.9-NA months, n=10) in PIK3CA Mt/EGFR Mt vs 9.1 months in PIK3CA Mt/EGFR wild-type patients (95% CI 4.4-NA months, p=0.87, n=16) (Figure 1C). Likewise, median OS was 8.9 months in PIK3CA Mt/KRAS Mt (95% CI 7.8-NA months, n=16) compared to 9.1 months in PIK3CA Mt/KRAS wild-type patients (95% CI 4.4-NA months, p=0.40, n=16) (Figure 1D).
Predictive impact of concurrent PIK3CA mutations on EGFR TKI therapy in EGFR-mutant lung cancers
No difference in best objective response to EGFR TKI monotherapy was noted between patients with EGFR single-Mt (n=29) and double-Mt (n=6) lung adenocarcinomas (18/29, 62% vs 5/6, 83%; p=0.80). Similarly, there was no difference in time to best response: 2.6 months (range 0.9-8.3 months) vs 1.9 months (range 0.7-13.6 months, p=0.65). Receipt of local therapy (radiation or surgery) for oligoprogressive disease did not differ between groups (p=0.99). These outcomes are summarized in Table 2.
Table 2.
Impact of PIK3CA Mutations on EGFR Tyrosine Kinase Inhibition (TKI) in patients with EGFR-mutant lung cancers.
| OUTCOMES WITH EGFR TKI THERAPY
| |||
|---|---|---|---|
| EGFR single-Mt | EGFR double-Mt | p value | |
| TKI analysis | N=32 | N=10 | |
|
| |||
| Duration of Therapy | |||
| Median Duration, 95% CI (months) | 14.5 (12.6-NA) | 14.6 (7.1-NA) | 0.65 |
| 6 months on therapy | 91% | 100% | |
| 1 year on therapy | 68% | 67% | |
| 2 years on therapy | 34% | 0% | |
| 3 years on therapy | 34% | 0% | |
|
| |||
| Local Therapy for Oligoprogressive Disease During Therapy | 8 (25%) | 2 (20%) | 0.99 |
|
| |||
| RECIST 1.1 analysis | N=29 | N=6 | |
|
| |||
| Best Response | |||
| CR | 2 (7%) | 0 | 0.80 |
| PR | 16 (55%) | 5 (83%) | |
|
| |||
| Time to Progression (TTP) | |||
| Median TTP, 95% CI (months) | 11.1 (7.5-18.4) | 7.8 (7.1-NA) | 0.84 |
| 6 months progression-free | 76% | 83% | |
| 1 year progression-free | 42% | 44% | |
| 2 years progression-free | 20% | 0% | |
| 3 years progression-free | 7% | 0% | |
Abbreviations: TKI, tyrosine kinase inhibitor; Mt, mutation; RECIST, Response Evaluation Criteria In Solid Tumors; CR, complete response; PR, partial response; TTP, time to progression
There was no difference in TTP between EGFR single-Mt and double-Mt patients: 11.1 months (95% CI 7.5-18.4 months) vs 7.8 months (95% CI 7.1-NA months, p=0.84). No difference in median duration of TKI monotherapy was noted (Figure 2): 14.5 months (95% CI 12.6-NA months) vs 14.6 months (95% CI 7.1-NA months, p=0.65).
Figure 2.
Duration of EGFR TKI Therapy for patients with EGFR single vs double-mutant lung cancers.
Subgroups with EGFR exon 19 deletions and L858R mutations were examined separately and did not reveal any significant differences (p>0.05) in TTP or duration of TKI monotherapy between single-Mt and double-Mt patients.
DISCUSSION
We demonstrate that the presence of a concurrent PIK3CA mutation in patients with EGFR-mutant and KRAS-mutant lung adenocarcinomas confers a distinct natural history. Median OS was substantially decreased compared to patients whose tumors did not harbor PIK3CA mutations. To our knowledge, this paper represents the largest series of oncogene-driven lung cancers with concurrent PIK3CA mutations to date, and is the first to demonstrate this phenomenon within specific genomic subsets of lung cancers. While the importance of single PIK3CA mutations as lung cancer drivers relative to other established oncogenes remains in question, this data argues that tumors with concurrent PIK3CA mutations represent a unique subset within EGFR- and KRAS-mutant lung cancers.
The data presented here is consistent with what is expected based on the known biological significance of PIK3CA in lung cancers and its cooperative role in mitogenic signaling pathways. In a genetically-engineered KRAS G12D-mutant murine lung cancer model, tumorigenesis was reliant on binding of the p110α subunit of PI3K to KRAS.11 Abrogation of this interaction resulted in a substantial reduction in lung tumor formation. In addition, given that PIK3CA H1074R was sufficient for the development of lung adenocarcinomas in vivo in a separate model,12 this raises the question of the impact of a concurrent PIK3CA mutation on oncogenesis within KRAS-mutant lung cancers. In BRAF-mutant lung adenocarcinoma mouse models, the concomitant expression of BRAF V600E and PIK3CA H1047R resulted in a dramatic increase in tumor burden compared to BRAF V600E alone, emphasizing the malignant synergy observed with co-existing mutations.13
Importantly, our series also represents the first systematic evaluation of the predictive value of PIK3CA mutations on EGFR TKI therapy in EGFR-mutant lung cancers. Preclinical studies have postulated that PIK3CA mutations could abrogate the effectiveness of EGFR TKIs as the presence of a PIK3CA mutation results in decreased response and diminished apoptosis with gefitinib in an EGFR-mutant lung cancer model.9 Furthermore, re-establishing PI3K pathway activity has been described as a mechanism of acquired resistance in EGFR-mutant lung cancers via the acquisition of PIK3CA mutations14 or activation of other receptor tyrosine kinases (e.g. MET amplification).15 In HER2-positive breast cancer, the presence of a PIK3CA mutation has been associated with lower pathologic complete response rates16 and inferior disease-free survival after neoadjuvant HER2-directed therapy.17 However, we found no evidence that a concurrent PIK3CA mutation impacts the outcomes of EGFR TKI therapy, including objective response, time to response, TTP, and duration of EGFR TKI therapy in patients with EGFR-mutant lung cancers. Although the small size of this retrospective study limits definitive conclusions, these data support the conclusion that the decision regarding whether to initiate EGFR TKI therapy in the clinic should be unaffected by the presence of a concurrent PIK3CA mutation.
Despite the absence of significant differences in outcomes with EGFR TKI monotherapy, several observations can be made. While all double-mutant patients (EGFR-Mt/PIK3CA-Mt) came off therapy within two years, about a third of single-mutant patients (EGFR-Mt/PIK3CA wild-type) remained on treatment for an additional year. In addition, complete responses were only noted in single-mutants. This highlights the need to confirm these data prospectively in ongoing and future trials of EGFR-directed therapies by including comprehensive molecular genomic profiling.
CONCLUSION
Concurrent PIK3CA mutations are poor prognostic factors within independent groups of EGFR- and KRAS-mutant lung adenocarcinomas. There was no evidence that clinical outcomes with EGFR TKI monotherapy in EGFR-mutant lung cancers are affected by PIK3CA status. These data illustrate that mutations in PIK3CA can act as biologically relevant driver oncogenes but do not negate the therapeutic benefits of molecular therapies in patients with concurrent targetable oncogenes.
Supplementary Material
Acknowledgments
DISCLAIMERS/FUNDING
Juliana Eng: none
Kaitlin M. Woo: none
Camelia S. Sima: Employment – Genentech
Andrew Plodkowski: none
Matthew D. Hellmann: none
Jamie Chaft: none
Mark G. Kris: Consultancy – AstraZeneca, Genentech, Novartis, Daiichi Sankyo; Grants – Pfizer, Puma; Payment for lectures – Genentech, Pfizer
Maria Arcila: Payment for lectures – AstraZeneca
Marc Ladanyi: Funding for research on which this article is based – National Institutes of Health
Alexander Drilon: Consultancy – Exelixis; Grands – ASCO CDA; Payment for lectures – Genentech; Royalties – Lippincott Pocket Oncology
References
- 1.Comprehensive molecular profiling of lung adenocarcinoma. Nature. 2014;511:543–550. doi: 10.1038/nature13385. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Zhao L, Vogt PK. Class I PI3K in oncogenic cellular transformation. Oncogene. 2008;27:5486–5496. doi: 10.1038/onc.2008.244. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Engelman JA. Targeting PI3K signalling in cancer: opportunities, challenges and limitations. Nature reviews Cancer. 2009;9:550–562. doi: 10.1038/nrc2664. [DOI] [PubMed] [Google Scholar]
- 4.Chaft JE, Arcila ME, Paik PK, et al. Coexistence of PIK3CA and other oncogene mutations in lung adenocarcinoma-rationale for comprehensive mutation profiling. Molecular cancer therapeutics. 2012;11:485–491. doi: 10.1158/1535-7163.MCT-11-0692. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Scheffler M, Bos M, Gardizi M, et al. PIK3CA mutations in non-small cell lung cancer (NSCLC): Genetic heterogeneity, prognostic impact and incidence of prior malignancies. Oncotarget. 2014 doi: 10.18632/oncotarget.2834. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Kawano O, Sasaki H, Endo K, et al. PIK3CA mutation status in Japanese lung cancer patients. Lung cancer. 2006;54:209–215. doi: 10.1016/j.lungcan.2006.07.006. [DOI] [PubMed] [Google Scholar]
- 7.Kris MG, Johnson BE, Berry LD, et al. Using multiplexed assays of oncogenic drivers in lung cancers to select targeted drugs. JAMA. 2014;311:1998–2006. doi: 10.1001/jama.2014.3741. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Engelman JA, Janne PA, Mermel C, et al. ErbB-3 mediates phosphoinositide 3-kinase activity in gefitinib-sensitive non-small cell lung cancer cell lines. Proceedings of the National Academy of Sciences of the United States of America. 2005;102:3788–3793. doi: 10.1073/pnas.0409773102. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Engelman JA, Mukohara T, Zejnullahu K, et al. Allelic dilution obscures detection of a biologically significant resistance mutation in EGFR-amplified lung cancer. The Journal of clinical investigation. 2006;116:2695–2706. doi: 10.1172/JCI28656. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Wang L, Hu H, Pan Y, et al. PIK3CA mutations frequently coexist with EGFR/KRAS mutations in non-small cell lung cancer and suggest poor prognosis in EGFR/KRAS wildtype subgroup. PloS one. 2014;9:e88291. doi: 10.1371/journal.pone.0088291. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Gupta S, Ramjaun AR, Haiko P, et al. Binding of ras to phosphoinositide 3-kinase p110alpha is required for ras-driven tumorigenesis in mice. Cell. 2007;129:957–968. doi: 10.1016/j.cell.2007.03.051. [DOI] [PubMed] [Google Scholar]
- 12.Engelman JA, Chen L, Tan X, et al. Effective use of PI3K and MEK inhibitors to treat mutant Kras G12D and PIK3CA H1047R murine lung cancers. Nature medicine. 2008;14:1351–1356. doi: 10.1038/nm.1890. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Trejo CL, Green S, Marsh V, et al. Mutationally activated PIK3CA(H1047R) cooperates with BRAF(V600E) to promote lung cancer progression. Cancer research. 2013;73:6448–6461. doi: 10.1158/0008-5472.CAN-13-0681. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Sequist LV, Waltman BA, Dias-Santagata D, et al. Genotypic and histological evolution of lung cancers acquiring resistance to EGFR inhibitors. Science translational medicine. 2011;3:75ra26. doi: 10.1126/scitranslmed.3002003. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Yu HA, Arcila ME, Rekhtman N, et al. Analysis of tumor specimens at the time of acquired resistance to EGFR-TKI therapy in 155 patients with EGFR-mutant lung cancers. Clinical cancer research : an official journal of the American Association for Cancer Research. 2013;19:2240–2247. doi: 10.1158/1078-0432.CCR-12-2246. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Loibl S, von Minckwitz G, Schneeweiss A, et al. PIK3CA mutations are associated with lower rates of pathologic complete response to anti-human epidermal growth factor receptor 2 (her2) therapy in primary HER2-overexpressing breast cancer. J Clin Oncol. 2014;32:3212–3220. doi: 10.1200/JCO.2014.55.7876. [DOI] [PubMed] [Google Scholar]
- 17.Cizkova M, Dujaric ME, Lehmann-Che J, et al. Outcome impact of PIK3CA mutations in HER2-positive breast cancer patients treated with trastuzumab. British journal of cancer. 2013;108:1807–1809. doi: 10.1038/bjc.2013.164. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.