Current Role at Stanford
Staff Scientist - Computational Biology
Education & Certifications
M.S., Northeastern University, Bioinformatics (2010)
Ph.D., Memorial Sloan Kettering Cancer Center, Cornell University, Computational Biology (2019)
Validation and clinical implementation of MSK-ACCESS, an ultra-deep sequencing assay for noninvasive somatic mutation profiling
AMER ASSOC CANCER RESEARCH. 2020: 29–30
View details for Web of Science ID 000537848000030
MSI detection in plasma cfDNA: MSI as a marker of disease burden
AMER ASSOC CANCER RESEARCH. 2020: 28–29
View details for Web of Science ID 000537848000029
Germline-focussed analysis of tumour-only sequencing: recommendations from the ESMO Precision Medicine Working Group
ANNALS OF ONCOLOGY
2019; 30 (8): 1221–31
It is increasingly common in oncology practice to perform tumour sequencing using large cancer panels. For pathogenic sequence variants in cancer susceptibility genes identified on tumour-only sequencing, it is often unclear whether they are of somatic or constitutional (germline) origin. There is wide-spread disparity regarding both the extent to which systematic 'germline-focused analysis' is performed upon tumour sequencing data and for which variants follow-up analysis of a germline sample is performed. Here we present analyses of paired sequencing data from 17,152 cancer samples, in which 1494 pathogenic sequence variants were identified across 65 cancer susceptibility genes. From these analyses, the European Society of Medical Oncology Precision Medicine Working Group Germline Subgroup have generated (i) recommendations regarding germline-focused analyses of tumour-only sequencing data, (ii) indications for germline follow-up testing and (iii) guidance on patient information-giving and consent.
View details for DOI 10.1093/annonc/mdz136
View details for Web of Science ID 000493072600007
View details for PubMedID 31050713
Tumour lineage shapes BRCA-mediated phenotypes
2019; 571 (7766): 576-+
Mutations in BRCA1 and BRCA2 predispose individuals to certain cancers1-3, and disease-specific screening and preventative strategies have reduced cancer mortality in affected patients4,5. These classical tumour-suppressor genes have tumorigenic effects associated with somatic biallelic inactivation, although haploinsufficiency may also promote the formation and progression of tumours6,7. Moreover, BRCA1/2-mutant tumours are often deficient in the repair of double-stranded DNA breaks by homologous recombination8-13, and consequently exhibit increased therapeutic sensitivity to platinum-containing therapy and inhibitors of poly-(ADP-ribose)-polymerase (PARP)14,15. However, the phenotypic and therapeutic relevance of mutations in BRCA1 or BRCA2 remains poorly defined in most cancer types. Here we show that in the 2.7% and 1.8% of patients with advanced-stage cancer and germline pathogenic or somatic loss-of-function alterations in BRCA1/2, respectively, selective pressure for biallelic inactivation, zygosity-dependent phenotype penetrance, and sensitivity to PARP inhibition were observed only in tumour types associated with increased heritable cancer risk in BRCA1/2 carriers (BRCA-associated cancer types). Conversely, among patients with non-BRCA-associated cancer types, most carriers of these BRCA1/2 mutation types had evidence for tumour pathogenesis that was independent of mutant BRCA1/2. Overall, mutant BRCA is an indispensable founding event for some tumours, but in a considerable proportion of other cancers, it appears to be biologically neutral-a difference predominantly conditioned by tumour lineage-with implications for disease pathogenesis, screening, design of clinical trials and therapeutic decision-making.
View details for DOI 10.1038/s41586-019-1382-1
View details for Web of Science ID 000477016700073
View details for PubMedID 31292550
View details for PubMedCentralID PMC7048239
Ampullary cancer: Evaluation of somatic and germline genetic alterations and association with clinical outcomes
2019; 125 (9): 1441–48
Ampullary carcinoma (AC) is a rare gastrointestinal cancer. Pathogenic germline alterations (PGAs) in BRCA2 and potentially targetable somatic alterations (SAs) in ERBB2 and ELF3 have been previously described in AC. Memorial Sloan Kettering Cancer Center has implemented an opt-in strategy for germline testing (GT) and somatic testing (ST) for patients with AC to further evaluate the spectrum of PGAs and SAs.Forty-five patients with pathologically confirmed AC prospectively consented with the Memorial Sloan Kettering Integrated Mutation Profiling of Actionable Cancer Targets (MSK-IMPACT) test (410-468 genes). A subset of the cohort (23 of the 45 patients) also consented to GT with MSK-IMPACT (76-88 genes). Germline data for 21 of the remaining 22 patients who had not consented to GT were obtained in a de-identified fashion without clinical correlation. Clinicopathologic features, treatment histories, and survival data for consenting patients were collected and analyzed.Pancreaticobiliary, intestinal, and mixed features of the 2 types were the primary pathologic subtypes of AC identified in this cohort. No difference in median overall survival was found between pathologic subtypes. Eight of 44 patients (18%) were identified as harboring pathogenic mutations in BRCA2, ATM, RAD50, and MUTYH. In addition, this study found a wide spectrum of SAs in genes such as KRAS, MDM2, ERBB2, ELF3, and PIK3CA. Two patients in the cohort underwent SA-targeted therapy, and 1 had a partial radiographic response.Mutations in multiple somatic and germline genes were identified in this cohort. Significantly, actionable targets were identified in the tumors, and broader testing for PGAs and SAs should be considered for all patients with AC.
View details for DOI 10.1002/cncr.31951
View details for Web of Science ID 000465035900010
View details for PubMedID 30620386
View details for PubMedCentralID PMC6467723
Understanding Inherited Risk in Unselected Newly Diagnosed Patients With Endometrial Cancer
JCO PRECISION ONCOLOGY
Mutations in DNA mismatch repair (MMR) genes and PTEN, diagnostic of Lynch and Cowden syndromes, respectively, represent the only established inherited predisposition genes in endometrial cancer to date. The prevalence of other cancer predisposition genes remains unclear. We sought the prevalence of pathogenic germline variants in unselected patients with endometrial cancer attending for surgical consultation.Patients were prospectively consented (4/2016-5/2017) to an IRB-approved protocol of tumor-normal sequencing via a custom next-generation sequencing panel (MSK-IMPACT) with return of germline results for >75 cancer predisposition genes. Tumors were assessed for microsatellite instability (MSI). Per institutional standards, all tumors underwent Lynch syndrome screening via IHC for MMR proteins.Of 156 patients who consented to germline genetic testing, 118 (76%) had stage I disease. Tumors were endometrioid in 104 (67%), of which 60 (58%) were grade 1. Twenty-four pathogenic germline variants were identified in 22 patients (14%)-7 (4.5%) with highly penetrant cancer syndromes and 15 (9.6%) with variants in moderate-, low-penetrance, or recessive genes. Of these, 5 (21%) were in Lynch syndrome genes (2 MSH6, 2 PMS2, and 1 MLH1). All 5 tumors had concordant IHC staining; 2 (40%) were definitively MSI-high by next-generation sequencing. One patient had a known BRCA1 mutation; 1 had SMARCA4 deletion. The remaining 17 variants (71%) were incremental findings in moderate- and low-penetrance variants or genes associated with recessive disease.In unselected patients with predominantly low-risk, early-stage endometrial cancer, germline multi-gene panel testing identifies cancer predisposition gene variants in 14%. This finding may have implications for future cancer screening and risk-reduction recommendations. Universal IHC screening for Lynch syndrome successfully identifies the majority (71%) of high-penetrance germline mutations.
View details for DOI 10.1200/PO.18.00338
View details for Web of Science ID 000493089500001
View details for PubMedID 32775946
View details for PubMedCentralID PMC7409950
Analysis of the Prevalence of Microsatellite Instability in Prostate Cancer and Response to Immune Checkpoint Blockade
2019; 5 (4): 471–78
The anti-programmed cell death protein 1 (PD-1) antibody pembrolizumab is approved by the US Food and Drug Administration for the treatment of microsatellite instability-high (MSI-H) or mismatch repair-deficient (dMMR) solid tumors, but the prevalence of MSI-H/dMMR prostate cancer and the clinical utility of immune checkpoint blockade in this disease subset are unknown.To define the prevalence of MSI-H/dMMR prostate cancer and the clinical benefit of anti-PD-1/programmed cell death 1 ligand 1 (PD-L1) therapy in this molecularly defined population.In this case series, 1551 tumors from 1346 patients with prostate cancer undergoing treatment at Memorial Sloan Kettering Cancer Center were prospectively analyzed using a targeted sequencing assay from January 1, 2015, through January 31, 2018. Patients had a diagnosis of prostate cancer and consented to tumor molecular profiling when a tumor biopsy was planned or archival tissue was available. For each patient, clinical outcomes were reported, with follow-up until May 31, 2018.Tumor mutation burden and MSIsensor score, a quantitative measure of MSI, were calculated. Mutational signature analysis and immunohistochemistry for MMR protein expression were performed in select cases.Among the 1033 patients who had adequate tumor quality for MSIsensor analysis (mean [SD] age, 65.6 [9.3] years), 32 (3.1%) had MSI-H/dMMR prostate cancer. Twenty-three of 1033 patients (2.2%) had tumors with high MSIsensor scores, and an additional 9 had indeterminate scores with evidence of dMMR. Seven of the 32 MSI-H/dMMR patients (21.9%) had a pathogenic germline mutation in a Lynch syndrome-associated gene. Six patients had more than 1 tumor analyzed, 2 of whom displayed an acquired MSI-H phenotype later in their disease course. Eleven patients with MSI-H/dMMR castration-resistant prostate cancer received anti-PD-1/PD-L1 therapy. Six of these (54.5%) had a greater than 50% decline in prostate-specific antigen levels, 4 of whom had radiographic responses. As of May 2018, 5 of the 6 responders (5 of 11 total [45.5%]) were still on therapy for as long as 89 weeks.The MSI-H/dMMR molecular phenotype is uncommon yet therapeutically meaningful in prostate cancer and can be somatically acquired during disease evolution. Given the potential for durable responses to anti-PD-1/PD-L1 therapy, these findings support the use of prospective tumor sequencing to screen all patients with advanced prostate cancer for MSI-H/dMMR. Because not all patients with the MSI-H/dMMR phenotype respond, further studies should explore mechanisms of resistance.
View details for DOI 10.1001/jamaoncol.2018.5801
View details for Web of Science ID 000464173600007
View details for PubMedID 30589920
View details for PubMedCentralID PMC6459218
Microsatellite Instability Is Associated With the Presence of Lynch Syndrome Pan-Cancer
JOURNAL OF CLINICAL ONCOLOGY
2019; 37 (4): 286-+
Microsatellite instability (MSI) and/or mismatch repair deficiency (MMR-D) testing has traditionally been performed in patients with colorectal (CRC) and endometrial cancer (EC) to screen for Lynch syndrome (LS)-associated cancer predisposition. The recent success of immunotherapy in high-frequency MSI (MSI-H) and/or MMR-D tumors now supports testing for MSI in all advanced solid tumors. The extent to which LS accounts for MSI-H across heterogeneous tumor types is unknown. Here, we establish the prevalence of LS across solid tumors according to MSI status.MSI status was determined using targeted next-generation sequencing, with tumors classified as MSI-H, MSI-indeterminate, or microsatellite-stable. Matched germline DNA was analyzed for mutations in LS-associated mismatch repair genes ( MLH1, MSH2, MSH6, PMS2, EPCAM). In patients with LS with MSI-H/I tumors, immunohistochemical staining for MMR-D was assessed.Among 15,045 unique patients (more than 50 cancer types), LS was identified in 16.3% (53 of 326), 1.9% (13 of 699), and 0.3% (37 of 14,020) of patients with MSI-H, MSI-indeterminate, and microsatellite-stable tumors, respectively ( P < .001). Among patients with LS with MSI-H/I tumors, 50% (33 of 66) had tumors other than CRC/EC, including urothelial, prostate, pancreas, adrenocortical, small bowel, sarcoma, mesothelioma, melanoma, gastric, and germ cell tumors. In these patients with non-CRC/EC tumors, 45% (15 of 33) did not meet LS genetic testing criteria on the basis of personal/family history. Immunohistochemical staining of LS-positive MSI-H/I tumors demonstrated MMR-D in 98.2% (56 of 57) of available cases.MSI-H/MMR-D is predictive of LS across a much broader tumor spectrum than currently appreciated. Given implications for cancer surveillance and prevention measures in affected families, these data support germline genetic assessment for LS for patients with an MSI-H/MMR-D tumor, regardless of cancer type or family cancer history.
View details for DOI 10.1200/JCO.18.00283
View details for Web of Science ID 000457744000005
View details for PubMedID 30376427
View details for PubMedCentralID PMC6553803
Clinical Utility of Prospective Molecular Characterization in Advanced Endometrial Cancer
CLINICAL CANCER RESEARCH
2018; 24 (23): 5939–47
Advanced-stage endometrial cancers have limited treatment options and poor prognosis, highlighting the need to understand genetic drivers of therapeutic vulnerabilities and/or prognostic predictors. We examined whether prospective molecular characterization of recurrent and metastatic disease can reveal grade and histology-specific differences, facilitating enrollment onto clinical trials.We integrated prospective clinical sequencing and IHC data with detailed clinical and treatment histories for 197 tumors, profiled by MSK-IMPACT from 189 patients treated at Memorial Sloan Kettering Cancer Center.Patients had advanced disease and high-grade histologies, with poor progression-free survival on first-line therapy (PFS1). When matched for histology and grade, the genomic landscape was similar to that of primary untreated disease profiled by TCGA. Using multiple complementary genomic and mutational signature-based methods, we identified patients with microsatellite instability (MSI), even when standard MMR protein IHC staining failed. Tumor and matched normal DNA sequencing identified rare pathogenic germline mutations in BRCA2 and MLH1. Clustering the pattern of DNA copy-number alterations revealed a novel subset characterized by heterozygous losses across the genome and significantly worse outcomes compared with other clusters (median PFS1 9.6 months vs. 17.0 and 17.4 months; P = 0.006). Of the 68% of patients harboring potentially actionable mutations, 27% were enrolled to matched clinical trials, of which 47% of these achieved clinical benefit.Prospective clinical sequencing of advanced endometrial cancer can help refine prognosis and aid treatment decision making by simultaneously detecting microsatellite status, germline predisposition syndromes, and potentially actionable mutations. A small overall proportion of all patients tested received investigational, genomically matched therapy as part of clinical trials.
View details for DOI 10.1158/1078-0432.CCR-18-0412
View details for Web of Science ID 000452374700018
View details for PubMedID 30068706
View details for PubMedCentralID PMC6279519
Mutational landscape of metastatic cancer revealed from prospective clinical sequencing of 10,000 patients
2017; 23 (6): 703-+
Tumor molecular profiling is a fundamental component of precision oncology, enabling the identification of genomic alterations in genes and pathways that can be targeted therapeutically. The existence of recurrent targetable alterations across distinct histologically defined tumor types, coupled with an expanding portfolio of molecularly targeted therapies, demands flexible and comprehensive approaches to profile clinically relevant genes across the full spectrum of cancers. We established a large-scale, prospective clinical sequencing initiative using a comprehensive assay, MSK-IMPACT, through which we have compiled tumor and matched normal sequence data from a unique cohort of more than 10,000 patients with advanced cancer and available pathological and clinical annotations. Using these data, we identified clinically relevant somatic mutations, novel noncoding alterations, and mutational signatures that were shared by common and rare tumor types. Patients were enrolled on genomically matched clinical trials at a rate of 11%. To enable discovery of novel biomarkers and deeper investigation into rare alterations and tumor types, all results are publicly accessible.
View details for DOI 10.1038/nm.4333
View details for Web of Science ID 000402768000013
View details for PubMedID 28481359
View details for PubMedCentralID PMC5461196
Clinical Relevance of KRAS-Mutated Subclones Detected with Picodroplet Digital PCR in Advanced Colorectal Cancer Treated with Anti-EGFR Therapy
CLINICAL CANCER RESEARCH
2015; 21 (5): 1087–97
KRAS mutations are predictive of nonresponse to anti-EGFR therapies in metastatic colorectal cancer (mCRC). However, only 50% of nonmutated patients benefit from them. KRAS-mutated subclonal populations nondetectable by conventional methods have been suggested as the cause of early progression. Molecular analysis technology with high sensitivity and precision is required to test this hypothesis.From two cohorts of patients with mCRC, 136 KRAS, NRAS, and BRAF wild-type tumors with sufficient tumor material to perform highly sensitive picodroplet digital PCR (dPCR) and 41 KRAS-mutated tumors were selected. All these patients were treated by anti-EGFR therapy. dPCR was used for KRAS or BRAF mutation screening and compared with qPCR. Progression-free survival (PFS) and overall survival (OS) were analyzed according to the KRAS-mutated allele fraction.In addition to the confirmation of the 41 patients with KRAS-mutated tumors, dPCR also identified KRAS mutations in 22 samples considered as KRAS wild-type by qPCR. The fraction of KRAS-mutated allele quantified by dPCR was inversely correlated with anti-EGFR therapy response rate (P < 0.001). In a Cox model, the fraction of KRAS-mutated allele was associated with worse PFS and OS. Patients with less than 1% of mutant KRAS allele have similar PFS and OS than those with wild-type KRAS tumors.This study suggests that patients with mCRC with KRAS-mutated subclones (at least those with a KRAS-mutated subclones fraction lower or equal to 1%) had a benefit from anti-EGFR therapies.
View details for DOI 10.1158/1078-0432.CCR-14-0983
View details for Web of Science ID 000351982800022
View details for PubMedID 25248381
Multiplex Picoliter-Droplet Digital PCR for Quantitative Assessment of DNA Integrity in Clinical Samples
2013; 59 (5): 815–23
Assessment of DNA integrity and quantity remains a bottleneck for high-throughput molecular genotyping technologies, including next-generation sequencing. In particular, DNA extracted from paraffin-embedded tissues, a major potential source of tumor DNA, varies widely in quality, leading to unpredictable sequencing data. We describe a picoliter droplet-based digital PCR method that enables simultaneous detection of DNA integrity and the quantity of amplifiable DNA.Using a multiplex assay, we detected 4 different target lengths (78, 159, 197, and 550 bp). Assays were validated with human genomic DNA fragmented to sizes of 170 bp to 3000 bp. The technique was validated with DNA quantities as low as 1 ng. We evaluated 12 DNA samples extracted from paraffin-embedded lung adenocarcinoma tissues.One sample contained no amplifiable DNA. The fractions of amplifiable DNA for the 11 other samples were between 0.05% and 10.1% for 78-bp fragments and ≤1% for longer fragments. Four samples were chosen for enrichment and next-generation sequencing. The quality of the sequencing data was in agreement with the results of the DNA-integrity test. Specifically, DNA with low integrity yielded sequencing results with lower levels of coverage and uniformity and had higher levels of false-positive variants.The development of DNA-quality assays will enable researchers to downselect samples or process more DNA to achieve reliable genome sequencing with the highest possible efficiency of cost and effort, as well as minimize the waste of precious samples.
View details for DOI 10.1373/clinchem.2012.193409
View details for Web of Science ID 000321547700017
View details for PubMedID 23403697
Propagation of Uncertainty in Bayesian Diagnostic Test Interpretation
SOUTHERN MEDICAL JOURNAL
2012; 105 (9): 452–59
Bayesian interpretation of diagnostic test results usually involves point estimates of the pretest probability and the likelihood ratio corresponding to the test result; however, it may be more appropriate in clinical situations to consider instead a range of possible values to express uncertainty in the estimates of these parameters. We thus sought to demonstrate how uncertainty in sensitivity, specificity, and disease pretest probability can be accommodated in Bayesian interpretation of diagnostic testing.We investigated three questions: How does uncertainty in the likelihood ratio propagate to the posttest probability range, assuming a point estimate of pretest probability? How does uncertainty in the sensitivity and specificity of a test affect uncertainty in the likelihood ratio? How does uncertainty propagate when present in both the pretest probability and the likelihood ratio?Propagation of likelihood ratio uncertainty depends on the pretest probability and is more prominent for unexpected test results. Uncertainty in sensitivity and specificity propagates into the calculation of likelihood ratio prominently as these parameters approach 100%; even modest errors of ± 10% caused dramatic propagation. Combining errors of ± 20% in the pretest probability and in the likelihood ratio exhibited modest propagation to posttest probability, suggesting a realistic target range for clinical estimations.The results provide a framework for incorporating ranges of uncertainty into Bayesian reasoning. Although point estimates simplify the implementation of Bayesian reasoning, it is important to recognize the implications of error propagation when ranges are considered in this multistep process.
View details for DOI 10.1097/SMJ.0b013e3182621a2c
View details for Web of Science ID 000308669000002
View details for PubMedID 22948322
View details for PubMedCentralID PMC6785978