PIK3CA Mutations are Common in Many Tumor Types and are Often Associated With Other Driver Mutations.

RESEARCH ARTICLE

PIK3CA Mutations are Common in Many Tumor Types and are Often Associated With Other Driver Mutations Matthew D. Stachler, MD, PhD, Elizabeth M. Rinehart, MD, Elizabeth Garcia, PhD, and Neal I. Lindeman, MD

Objectives: Genotyping clinical cancer specimens determines a fuller spectrum of mutations that an individual tumor harbors, and thus provides better insight into its molecular pathogenesis. Using genotyping data collected during routine clinical care our objective was to better determine the genomic landscape associated with PIK3CA mutations since much interest has been placed on PIK3CA targeted therapy. Methods: We performed multiplexed tumor genotyping within our CLIA-certified clinical laboratory on all consenting cancer patients who presented to the Brigham and Women’s Hospital/ Dana-Farber Cancer Center, regardless of histologic subtype. A total of 3252 cancers were genotyped for 471 mutations in 41 cancer-associated genes (including 23 mutations in PIK3CA), using a PCR-mass spectrometry assay. Results: A total of 288 (9%) samples contained a mutation in PIK3CA, involving 25 different primary sites. In 117 (41%) cases, the PIK3CA mutation was found with at least 1 other mutation, many known oncogenic drivers, while only 7% of the non-PIK3CA mutated cases, when comparing like tumor types, had >1 mutation (P < 0.0001). Breast cancers had the highest rate of PIK3CA mutations (34%), which correlated with estrogen receptor + status (P = 0.0002). Conclusions: These findings suggest that PIK3CA mutations may be a relatively late event and may function primarily in a supporting/modifying role, and not as a primary driver of oncogenesis. Although further studies are needed, our observations during clinical tumor genotyping suggest that when other pro-oncogenic pathways are mutated along with PIK3CA, then, PIK3CA inhibition alone may not be effective and combination therapy may be warranted. Key Words: cancer, genotyping, PIK3CA, mutations, targeted therapy (Appl Immunohistochem Mol Morphol 2016;24:313–319)

Received for publication December 19, 2014; accepted February 12, 2015. From the Department of Pathology, Center for Advanced Molecular Diagnostics, Brigham and Women’s Hospital, Boston, MA. The authors declare no conflict of interest. Reprints: Matthew D. Stachler, MD, PhD, Department of Pathology, Shapiro 5, Room 016, Center for Advanced Molecular Diagnostics, Brigham and Women’s Hospital, 75 Francis Street, Boston, MA 02115 (e-mail: [email protected]). Copyright r 2015 Wolters Kluwer Health, Inc. All rights reserved.

Appl Immunohistochem Mol Morphol

T

he phosphoinositide 3-kinases have been known to have oncogenic potential for close to 15 years.1 Since then, it has been shown that the p110a catalytic subunit (PIK3CA) is one of the most frequently mutated genes in cancer, having been reported in B12% of all cancers in the Catalogue of Somatic Mutations in Cancer (COSMIC) database. PIK3CA is composed of multiple domains,2 all of which have been associated with gain-of-function mutations.3,4 Most mutations occur in specific “hotspots” including amino acid E542 and E545 of the helical domain and H1047 in the kinase domain.5,6 While significant work has been performed to determine the frequency and scope of PIK3CA mutations, less work has been done identifying the background mutational profile of cancers harboring these mutations. As the majority of the mutations are missense and occur in hotspots, broad genotyping technologies that can analyze many genes are well suited to investigate samples for PIK3CA mutations.7 The widespread nature of PIK3CA mutations has generated significant interest in developing targeted therapy. While some studies have shown limited success with partial response rates usually around 15% to 30% in patients with PIK3CA mutations, most have shown a lack of prolonged progression-free survival.8–11 Given limited efficacy with monotherapy of PI3-K pathway–directed agents, there is increasing interest in use of these agents as part of combination strategies. Development of such strategies, however, will be greatly facilitated by expanded knowledge of the genomic context in which PIK3CA mutations occur and how this context differs across distinct classes of cancer. Until recently, widespread molecular characterization of patient samples was difficult to perform in clinical molecular labs; entities that have traditionally performed focused single-gene assays. Today’s convergence of expanding insight into the genomic underpinnings of cancer and the increasing availability of economically feasible and massively multiplexed genomic profiling make possible more comprehensive routine molecular characterization of clinical samples. We have implemented Profile, a program to perform multiplexed genotyping assays routinely in the workflow of our cancer center. Importantly, our testing is performed on all consenting patients with sufficient cancer tissue to test, without selection by cancer type, stage, or any other clinical, socioeconomic, or pathologic variables. This enables a broad assessment of mutational status across the entire spectrum of cancers seen in our cancer center. The technology initially used in the Profile program, termed OncoMap, utilizes single-base extension chemistry and MALDI-TOF mass spectrophotometry (Agena, San

Volume 24, Number 5, May/June 2016

www.appliedimmunohist.com |

Copyright r 2016 Wolters Kluwer Health, Inc. All rights reserved.

313


Stachler et al

TABLE 1. Genes Included in the OncoMap Platform for Genotyping Genes Genotyped ABL1 AKT1 AKT2 APC BRAF CDK4 CDKN2A CSF1R CTNNB1 EGFR ERBB2

FGFR1 FGFR2 FGFR3 FLT3 GNA11 GNAQ GNAS HRAS IDH1 IDH2

JAK2 JAK3 KIT KRAS MAP2K1 MET MLH1 MYC NPM1 NRAS

PDGFRA PIK3CA PIK3R1 PTEN RB1 RET SRC STK11 TP53 VHL

Diego, CA) to genotype 471 hotspots in 41 genes12,13 (Table 1). Included in this are 23 of the most common alterations in PIK3CA (Table 2). Here we report the PIK3CA mutational status of the first 3252 consecutive cancer cases genotyped with the goal of characterizing the number and type of concurrent mutations that are found in PIK3CA mutant tumors and comparison of the clinicopathologic status of select types of tumors. Through more detailed characterization of the molecular profile of PIK3CA mutant cancers, these data may improve the design, rationale, and outcome of PI3K-targeted therapy trials.

MATERIALS AND METHODS Acquisition of Patient Samples Beginning August 2011, all patients being treated for neoplastic disease at the Dana-Farber Cancer Institute)


and Brigham and Women’s Hospital (BWH) had been asked to consent for tumor genotyping (IRB11-104). For those cases for which patient consent had been documented, excess material from clinically obtained pathologic samples (formalin-fixed paraffin-embedded, or frozen) were retrieved and testing was performed in the Center for Advanced Molecular Diagnostics, BWH.

DNA Isolation After processing of specimens by the histopathology laboratory, H&E-stained slides were reviewed by a pathologist to determine adequacy and the area of highest tumor percentage. Inclusion criteria for samples were >30% malignant cell content, >3 mm in greatest linear dimension, and sufficient material to prepare ten 4 mm sections or five 1 mm tissue cores without exhausting the sample. The area of highest tumor percentage was then macrodissected. For large tumors, a sample of tumor was taken to the gross pathology laboratory for DNA isolation along with a frozen section slide to determine tumor percentage. Overall, approximately 90% of the total samples were formalin-fixed paraffin-embedded and 10% were frozen. DNA was isolated using a Qiasymphony (Qiagen, Valencia, CA) automated DNA extractor according to the manufacturer’s recommendations. DNA was quantified using SYBR-green–based dsDNA detection. Samples with >250 ng total DNA at a concentration of at least 5 ng/mL were genotyped. Approximately 54% of the patients with consent, requisitions, and a tissue block were deemed to have sufficient tissue for further processing. The failed samples were typically either small biopsies (especially in prostate or lung) or samples with very low tumor percentage such as postneoadjuvant samples.

TABLE 2. PIK3CA Mutations Identified by OncoMap Tumor Genotyping C.DNA Change*

Codon Change

Amino Acid Change

Exon

% in COSMICw

c.263G > A c.1035T > A c.1258T > C c.1616C > G c.1624G > A c.1624G > C c.1633G > A c.1633G > C c.1634A > C c.1634A > G c.1635G > C c.1635G > T c.1636C > A c.2102A > C c.3062A > G c.3129G > A c.3129G > T c.3139C > T c.3140A > G c.3140A > T c.3145G > A c.3145G > C c.3204_3205insA

CGA > CAA AAT > AAA TGT > CGT CCT > CGT GAA > AAA GAA > CAA GAG > AAG GAG > CAG GAG > GCG GAG > GGG GAG > GAC GAG > GAT CAG > AAG CAC > CCC TAC > TGC ATG > ATA ATG > ATT CAT > TAT CAT > CGT CAT > CTT GGT > AGT GGT > CGT insA

p.R88Q p.N345K p.C420R p.P539R p.E542K p.E542Q p.E545K p.E545Q p.E545A p.E545G p.E545D p.E545D p.Q546K p.H701P p.Y1021C p.M1043I p.M1043I p.H1047Y p.H1047R p.H1047L p.G1049S p.G1049R p.N1068fs*4

1 4 6 9 9 9 9 9 9 9 9 9 9 13 20 20 20 20 20 20 20 20 20

2.10 0.86 0.93 0.39 11.45 0.47 15.19 0.55 3.43 1.56 0.39 0.39 1.64 0.00 0.93 0.62 0.86 0.78 25.16 3.43 0.16 0.70 1.09

*Reference sequence NM_006218.2. wPercentage of total PIK3CA mutations with known c.DNA change listed in COSMIC.

314 | www.appliedimmunohist.com

Copyright

r

2015 Wolters Kluwer Health, Inc. All rights reserved.




TABLE 3. Tumor Location and Percent of Cases Containing a PIK3CA Mutation Total No. Cases Genotyped

Location Adrenal gland Anus Appendix Biliary tract Bone Breast Central nervous system Cervix Endometrium Esophagus Eye Female genital tract (unknown) Gastrointestinal tract (unknown) Hematopoietic and lymphoid tissue Heart Kidney Large intestine Liver Lung Mediastinum Meninges Ovary (and fallopian tube) Pancreas Parathyroid Penis Peritoneum Pituitary Pleura Prostate Salivary gland Skin Small intestine Soft tissue Spleen Stomach Testis Thymus Thyroid Upper aerodigestive tract Urinary tract Uterus (myometrium) Vagina Vulva

No. Cases With PIK3CA Mutations (%)

% in COSMIC*

15 12 11 13 10 342 160

0 0 0 3 1 116 12

(0) (0) (0) (23) (10) (34)w (8)

0 0 0 8 0 25 5

24 211 37 3 7

5 66 1 0 1

(21) (31)w (3) (0) (14)

13 22 6 1 NA

6 248

0 (0)

4

1 (0.4)

1

1 104 224 11 362 8 1 242

0 3 25 2 4 0 0 10

(0) (3) (11) (18) (1)w (0) (0) (4)w

NA 2 14 6 3 NA 1 9

51 8 5 19 1 90 149 20 103 44 170 2 70 19 11 132 65

1 0 0 1 0 0 1 1 3 1 3 0 0 0 0 1 7

(2) (0) (0) (5) (0) (0) (1) (5) (3)w (2) (2) (0) (0)w (0) (0) (1) (11)

3 0 29 0 3 1 2 0 11 6 2 NA 9 0 0 4 6

125 20 1 9

17 0 0 1

(14) (0) (0) (11)

17 NA 0 0

*Percentage of unique samples with a PIK3CA mutation by location listed in COSMIC as of July 10, 2013. wDifference between observed and COSMIC P < 0.05.

However, of those passing review, 95% gave adequate DNA.

Tumor Genotyping Our OncoMap genotyping platform utilizes mass spectrometry detection and 2 different chemistries (screening and confirmation). DNA was divided into 2 Copyright

r


PIK3CA Mutations in Various Tumor Types

TABLE 4. Number of Cases with >1 Mutation Identified

Samples with a PIK3CA mutation Samples without a PIK3CA mutation

Samples With Z2 Mutations

Total No. Samples

%

117

288

40.6

206

2964

6.95

aliquots with 1 undergoing whole genome amplification (WGA). A panel of 48 SNPs was performed on both aliquots to confirm the comparability between the 2 aliquots. The WGA aliquots were used in a first round screening test, using single-base extension chemistry (iPLEX) and MALDI-TOF mass spectrophotometry (Agena). All mutations detected by the iPLEX assay were then confirmed with a more specific (homogenous mass extension) method on the non-WGA amplified aliquot, as previously described.12–14 A total of 99.9% of samples with adequate DNA were successfully genotyped.

Analysis Interpretation was performed manually, guided by a computational algorithm based upon k-means clustering as previously described.12 The number of PIK3CA mutations found from each primary tumor site was quantified. The total number and types of mutations as well as the basic clinicopathologic characteristics of various tumor types were compared between PIK3CA-mutated and non-PIK3CA-mutated cases.

Statistical Analysis For categorical data, a Fisher exact test was used for statistical analysis (http://www.graphpad.com/quick calcs/). For noncategorical data, a Student t test was performed within an Excel spreadsheet. A value of P < 0.05 was considered statistically significant.

RESULTS A total of 3252 cancers were genotyped. The 23 PIK3CA mutations (Table 2) in our assay would have detected 73.1% of the total instances of PIK3CA mutations reported in the Sanger Institute Catalogue of Somatic Mutations in Cancer database (COSMIC) dataset.7 Of the 3252 cases genotyped, 1331 (41%) had at least 1 mutation and 323 (10%) had Z2 mutations. Within the tumors profiled, 288 (9%) samples had a mutation in PIK3CA consistent with the B12% prevalence reported in COSMIC, coupled with the limitation of the assay design, which was to detect the 23 most prevalent PIK3CA mutations. These mutations were found in 25 different primary tumor sites (Table 3). As part of the initial validation and standard positive controls, 29 oligonucleotides covering all of the PIK3CA variants were spiked into control DNA. For PIK3CA we achieved 100% sensitivity and specificity on all runs. In total, for all mutations, 432 of 439 assays detected the appropriate alteration. On all 7 assays that failed, it was www.appliedimmunohist.com |


315


Stachler et al


TABLE 5. Other Genes with Mutations Identified in Same Sample as PIK3CA Location

Genes (No. Times Identified)

Biliary tract Bone (sarcoma) Bone marrow (leukemia) Breast

None TP53(1) None AKT1 (1), CDKN2A (1), CTNNB1 (1), HRAS (1), JAK3 (1), KRAS (2), MET (3), TP53 (3), PTEN (1), RET (1), 2nd PIK3CA (4) BRAF (1), TP53 (3) ERBB2 (1), TP53 (1, SCC) CTNNB1 (15), FGFR2 (5), IDH1 (1), KRAS (14), MET (2), TP53 (5), PIK3R1 (4), PTEN (16), RB1 (1), 2nd PIK3CA (1) KRAS (1), 2nd PIK3CA (1) None AKT2 (1), APC (7), BRAF (3), CTNNB1 (3), KIT (1), KRAS (9), TP53 (5) APC (1), KRAS (2) KRAS (2) CTNNB1 (2), FGFR2 (2) None KRAS (1) None TP53 (1) KIT (1) None None CTNNB1 (1), FGFR3 (2), HRAS (1), KRAS (2) None

Central nervous system Cervix Endometrium Esophagus Kidney (RCC) Large intestine Liver (HCC) Lung Ovary Pancreas Peritoneum Prostate Skin (SCC) Small intestine (GIST) Soft tissue (sarcoma) Head and neck (SCC) Urinary tract (urothelial) Vulva (SCC)

Parentheses under Location denote nonadenocarcinomas. GIST indicates gastrointestinal stromal tumor; HCC, hepatocellular carcinoma; RCC, renal cell carcinoma; SCC, squamous cell carcinoma.

due to poor overall performance (failed run). In addition, PIK3CA status was confirmed on a sequencing platform for 6 samples. These results were 100% concordant (4 positive and 2 negative). Interestingly, in 117 (41%) cases the PIK3CA mutation was found with at least 1 other mutation, whereas only 7% of the non-PIK3CA-mutated cases (when comparing like tumor types) had >1 mutation (P < 0.0001) (Table 4). Generally, the additional mutations paralleled mutations that are commonly reported “driver” genes in the individual tumor histologic type. In colorectal adenocarcinoma (CRC), 72% (N = 18) of the cases containing a PIK3CA mutation also contained an additional mutation; 9 (50%) of these cases contained a KRAS mutation, 3 (17%) contained a BRAF mutation, and 3 (17%) of cases also contained a CTNNB1 mutation. Interestingly, 1 PIK3CA-mutated CRC contained a potentially targetable KIT mutation. For breast cancer, only 18 (16%) of the cases with PIK3CA mutations also contained an additional mutation; this included 4 (22%) with multiple mutations in PIK3CA and 17% with mutations in RAS (2 KRAS and 1 HRAS) genes. Forty-seven (71%) of the PIK3CA-mutated endometrial cancers had additional mutation(s), including many in potentially targetable pathways. Of these, 16 (34%) had a PTEN mutation, 15 (32%) had a CTNNB1 mutation, 14 (30%) had a KRAS or BRAF mutation, 5 (11%) had a mutation in FGFR2, and 3 (6%) had an additional mutation in PIK3CA or PIK3R1. Overall, concurrent mutations included 58 in the receptor tyrosine kinase-Ras pathway, 32 in WNT pathway, 24 in other PI3K genes, 21 in TP53, 2


in RB pathway, 2 in KIT, and 1 in IDH. The additional mutations found in PIK3CA-mutated cancers are summarized in Table 5. Breast cancers had the highest percentage of PIK3CA mutations. Fifty-one (27.6%) invasive ductal carcinomas, 12 (37.5%) invasive lobular carcinoma, and 34 (35.0%) invasive carcinomas with mixed features were positive for PIK3CA mutations. PIK3CA mutation status did not correlate with lymph node status in either ductal or lobular subtypes of breast cancer, where 22 (51%) and 57 (46%) cases had positive nodes and 21 (49%) and 68 (54%) had negative nodes for PIK3CA-mutated and nonPIK3CA-mutated ductal carcinoma cases, respectively (P = 0.59). However, mutation status did correlate with positive estrogen receptor expression in the ductal carcinoma cases (P = 0.0002), and trended toward correlation in lobular carcinoma (P = 0.13); data are summarized in Figure 1. PIK3CA mutations in both ductal and lobular carcinoma were predominately found in exon 20 (57% and 50% of total, respectively) and exon 9 (35% and 33% of total, respectively). However, 8% (N = 4) of the ductal carcinoma PIK3CA mutations were found in exon 4 (0% in lobular), whereas 17% (N = 2) of the lobular carcinoma PIK3CA mutations were found in exon 7 (0% on ductal). Exon location did not correlate with histologic grade. In endometrial cancers, PIK3CA mutations occurred in 50 (34.2%) endometrioid, 5 (18.5%) serous, 3 (37.5%) mixed type, and 3 (21.4%) carcinosarcoma endometrial cancers, which accounted for the second highest frequency of mutations. Exon 20 mutations tended to correlate with Copyright

r




A



200 180 160

Number of Cases

140 120

PIK3CA mutated Non-PIK3CA mutated

100 80 60 40 20 0 Other or unspecified

Ductal

Lobular

Ductal

B

Mixed-type

Mucinous

Lobular

C

P=0.43 16

100 90 80 70 60 50 40 30 20 10 0

14

Number of Cases

Number of Cases

P=0.6

10 8 6 4

0

E

P=0.0002

140

Node +

Node -

Node P=0.13

35 30

Number of Cases

120

Number of Cases

12

2 Node +

D

Non-epithelial tumors

100 80 60 40 20

25 20 15 10 5

0

0 ER +

ER -

ER +

ER -

FIGURE 1. PIK3CA mutations in breast cancer. A, Breakdown of mutational status in the different histologic types of breast cancer. B and C, PIK3CA and nodal status in ductal (B) and lobular (C) breast carcinoma. D and E, PIK3CA and estrogen receptor status in ductal (B) and lobular (C) breast carcinoma.

high-grade (grade 3) endometrioid adenocarcinomas compared with other PIK3CA mutations (P = 0.08). PIK3CA mutation status did not correlate with metastatic status or Copyright

r


nodal status. Concurrent PTEN mutations were identified in 16 (26%) endometrial cancers, all either endometrioid or mixed with an endometrioid component. www.appliedimmunohist.com |


317


Stachler et al

DISCUSSION The rapidly increasing number of genes that need to be tested, decreasing costs of multiplex testing, and the desire to accumulate data that are not only relevant for currently approved therapies but may also help guide placement into clinical trials is driving the switch from the traditional single-gene tests to multigene platforms. During this process, a tremendous amount of data are generated that can expand our understanding of cancer pathogenesis. Here we report our findings in relation to PIK3CA mutations, one of the most frequently mutated genes in cancer. By analyzing the mutation status of a wide variety of genes in an entire institutional cohort of cancer patients, we were able to determine which genes and how frequently they are concurrently mutated with PIK3CA across cancer types. Our data indicate PIK3CA mutations are found in a very wide variety of tumors and tend to occur with other mutations. Mutations in multiple different pathways, including many in driver genes, were observed, but there was not a significant association between PIK3CA mutations and any other specific mutation. Importantly, this finding may have major implications for targeted therapy, as many PI3K inhibitor trials are underway or being designed in a variety of cancers. Indeed, a simple search for “PIK3CA” on clinicaltrials.gov produces 47 studies (http://www.ClincialTrials.gov). If other prooncogenic pathways are commonly mutated along with PIK3CA, then PIK3CA inhibition alone may not be effective and combination therapy may be warranted. The underlying mechanism of this association is unknown, but suggests that either PIK3CA mutations are a later event occurring after other mutations have occurred or somehow lead to additional mutations. While we did not find a correlation with advanced stage disease in either breast or endometrial cancer, it is unknown how mutations in PIK3CA could lead to an increased frequency of mutations. Despite the many studies showing the oncogenic potential of PIK3CA activating mutations, it is possible they only play a supporting role and are not a primary driver in many cancers. Despite our finding of a lack of association with advanced disease, the fact that previous studies have shown PIK3CA mutations to be associated with advanced disease would seem to support the conclusion that PIK3CA mutations are later events.3,15,16 It is unknown why we did not observe a correlation between advanced stage of disease in breast and endometrial cancers as has been reported previously,3,15,16 as the majority of PIK3CA mutations are detectable with this platform. However, as this assay detects 23 of the most common mutations, rare mutations not in the assay design were likely missed. Sample sizes and subtypes of the individual cancers also varied between studies. All of the cases for this study were accrued from a single tertiary care center. It is possible that the referral nature of our institution led to testing more advanced cases than is typical as many lower stage cancers may be treated at local hospitals outside our institution. In addition, the



percentage of the types of cancer tested does not completely reflect the overall incidence of these cancers in the general population. However, the mutations (and percentages) seen within the individual types of cancers should not be altered. While we present a greatly expanded view of the genomic profile in cancers harboring PIK3CA mutations, our study is still somewhat limited. With the OncoMap platform we are only able to detect mutations in 41 other genes and do not have data on copy number events. With an expanded platform (such as next-generation sequencing) that is able to analyze more genes in a comprehensive manner, it is likely that many additional alterations would be identified. While many current trials only require the mutational status of the targeted gene be known, having an understanding of the other alterations will give possible insight into modes of resistance if the therapy fails. Going forward, dual (and even triple) targeted therapy will likely be utilized. Here it will be vital to know what other alterations are present to determine which other pathway to target. In summary, PIK3CA mutations are common throughout a wide variety of tumors. Tumors harboring PIK3CA mutations tend to have an increased number of additional mutations compared with non-PIK3CA-mutated cancers of the same type. These data could be valuable clinically as many of these mutations may lead to resistance to PIK3CA-targeted inhibition and suggests that combination therapy may be required when utilizing PIK3CA inhibitors. REFERENCES 1. Chang HW, Aoki M, Fruman D, et al. Transformation of chicken cells by the gene encoding the catalytic subunit of PI 3-kinase. Science. 1997;276:1848–1850. 2. Walker EH, Perisic O, Ried C, et al. Structural insights into phosphoinositide 3-kinase catalysis and signaling. Nature. 1999;402: 313–320. 3. Rudd ML, Price JC, Fogoros S, et al. A unique spectrum of somatic PIK3CA (p110alpha) mutations within primary endometrial carcinomas. Clin Cancer Res. 2011;17:1331–1340. 4. Gymnopoulos M, Elsliger MA, Vogt PK. Rare cancer-specific mutations in PIK3CA show gain of function. Proc Natl Acad Sci USA. 2007;104:5569–5574. 5. Samuels Y, Wang Z, Bardelli A, et al. High frequency of mutations of the PIK3CA gene in human cancers. Science. 2004;304:554. 6. Vogt PK, Kang S, Elsliger MA, et al. Cancer-specific mutations in phosphatidylinositol 3-kinase. Trends Biochem Sci. 2007;32:342–349. 7. Wellcome Trust Sanger Institute. COSMIC catalogue of somatic mutations in cancer. Available at: http://cancer.sanger.ac.uk/cancer genome/projects/cosmic/. Accessed January 2014. 8. Janku F, Wheler JJ, Naing A, et al. A PIK3CA mutations in advanced cancers: characteristics and outcomes. Oncotarget. 2012;3:1566–1575. 9. Janku F, Wheler JJ, Westin SN, et al. PIK3/AKT/mTOR inhibitors in patients with breast and gynecologic malignancies harboring PIK3CA mutations. J Clin Oncol. 2012;30:777–782. 10. Janku F, Wheler JJ, Naing A, et al. PIK3CA mutation H1047R is associated with response to PIK3CA/AKT/mTOR signaling pathway inhibitors in early-phase clinical trials. Cancer Res. 2013;73: 276–284. 11. Brana I, Siu L. Clinical development of phosphatidylinositol 3-kinase inhibitors for cancer treatment. BMC Med. 2012;10:161.

Copyright

r





12. MacConaill LE, Campbell CD, Kehoe SM, et al. Profiling critical cancer gene mutatinos in clinical tumor samples. PLoS One. 2009;4:e7887. 13. MacConail L, Garcia E, Shivdasani P, et al. Prospective enterpriselevel molecular genotyping of a cohort of cancer patients. J Mol Diagn. 2014;16:660–672. 14. Thomas RK, Baker AC, Debiasi RM, et al. High-throughput oncogene mutation profiling in human cancer. Nat Genet. 2007;39:347–351.

Copyright

r



15. Saal LH, Holm K, Maurer M, et al. PIK3CA mutations correlate with hormone receptors, node metastasis, and ERBB2, and are mutually exclusive with PTEN loss in human breast carcinoma. Cancer Res. 2005;65:2554–2559. 16. Catasus L, Gallardo A, Cuatrecasas M, et al. PIK3CA mutations in the kinase domain (exon 20) of uterine endometrial adenocarcinomas are associated with adverse prognostic parameters. Mod Pathol. 2008;21:131–139.

www.appliedimmunohist.com |


319

overgrowth disorders are caused by somatic mutations in PIK3CA.

Microparticle phenotypes are associated with driver mutations and distinct thrombotic risks in essential thrombocythemia.

Somatic PIK3CA mutations as a driver of sporadic venous malformations.

Mutations in HECW2 are associated with intellectual disability and epilepsy.

AKT pathway genes and amplifications of PIK3CA are associated with patterns of recurrence in gastric cancers.

Hotspot mutations in common oncogenes are infrequent in nasopharyngeal carcinoma.

PIK3CA mutations in meningioma.

Oncogenic PI3K mutations are as common as AKT1 and SMO mutations in meningioma.

ARMC5 mutations are common in familial bilateral macronodular adrenal hyperplasia.

Recurrent tissue-specific mtDNA mutations are common in humans.

Mutations of epigenetic regulatory genes are common in thymic carcinomas.

Germline mutations in MAP3K6 are associated with familial gastric cancer.

Functional chromatin features are associated with structural mutations in cancer.

GBM-associated mutations and altered protein expression are more common in young patients.

Microsatellite instability derived JAK1 frameshift mutations are associated with tumor immune evasion in endometrioid endometrial cancer.

SPG7 mutations are a common cause of undiagnosed ataxia.

Germline RECQL mutations are associated with breast cancer susceptibility.

Common cancer-associated PIK3CA activating mutations rarely occur in Langerhans cell histiocytosis.

SGCZ mutations are unlikely to be associated with myoclonus dystonia.

DMRT1 mutations are rarely associated with male infertility.

CACNA1H Mutations Are Associated With Different Forms of Primary Aldosteronism.

Somatic MAP2K1 Mutations Are Associated with Extracranial Arteriovenous Malformation.

SAMHD1 Gene Mutations Are Associated with Cerebral Large-Artery Atherosclerosis.

Are CHCHD10 mutations indeed associated with familial amyotrophic lateral sclerosis?