Laboratory Considerations for Releasing Next-Generation Sequencing Data to Patients

CONTEXT.—

Title 45, section 164.524 of the Code of Federal Regulations states that health care systems must provide patient health records upon that patient's request. For complex testing, such as next-generation sequencing (NGS), this raises questions related to what data should be released and the laboratory considerations regarding the release of this data.

OBJECTIVE.—

To describe the laboratory implications of releasing different NGS data files and the limitations for the clinical use of different NGS data files.

DESIGN.—

The College of American Pathologists workgroup, composed of laboratorians with expertise regarding NGS testing, reviewed pertinent literature, including title 45, section 164.524, and the Health and Human Services "Guidance on Individuals' Right to Access Health Information."

RESULT.—

From an accreditation standpoint, validation of NGS includes both the wet bench and data processing (bioinformatics) portions, and appropriately validated laboratory testing is required to ensure quality patient results. NGS testing generates intermediate data files that have not completed the fully validated process but are often kept by the laboratory. These files may be requested by patients, but most patients will not be aware of the test validation process and the limitations of data that have not gone through a fully validated process.

CONCLUSIONS.—

Laboratories should encourage patients to receive their health data and to help individuals understand the content, uses, and limitations of laboratory data they have requested or received. NGS data used in a nonvalidated manner should not be used for clinical purposes without confirmation by a clinically validated method.

Evaluation of the Aggressive-Variant Prostate Cancer Molecular Signature in Clinical Laboratory Improvement Amendments (CLIA) Environments

Aggressive-variant prostate cancers (AVPCs) are a subset of metastatic castrate-resistant prostate cancers (mCRPCs) characterized by defects in ≥ two of three of TP53, RB1, and PTEN (AVPCm), a profile linked to lineage plasticity, androgen indifference, and platinum sensitivity. Men with mCRPC undergoing biopsies for progression were assessed for AVPCm using immunohistochemistry (IHC), next-generation sequencing (NGS) of solid tumor DNA (stDNA), and NGS of circulating tumor DNA (ctDNA) assays in CLIA-certified labs. Biopsy characteristics, turnaround times, inter-reader concordance, and inter-assay concordance were assessed. AVPCm was detected in 13 (27%) patients via IHC, two (6%) based on stDNA, and seven (39%) based on ctDNA. The concordance of the IHC reads between pathologists was variable. IHC had a higher detection rate of AVPCm+ tumors with the shortest turnaround times. stDNA had challenges with copy number loss detection, limiting its detection rate. ctDNA detected the greatest proportion of AVPCm+ tumors but had a low tumor content in two thirds of patients. These data show the operational characteristics of AVPCm detection using various assays, and inform trial design using AVPCm as a criterion for patient selection or stratification.

Mismatch Repair and Microsatellite Instability Testing for Immune Checkpoint Inhibitor Therapy: Guideline From the College of American Pathologists in Collaboration With the Association for Molecular Pathology and Fight Colorectal Cancer

CONTEXT.—

The US Food and Drug Administration (FDA) approved immune checkpoint inhibitor therapy for patients with advanced solid tumors that have DNA mismatch repair defects or high levels of microsatellite instability; however, the FDA provided no guidance on which specific clinical assays should be used to determine mismatch repair status.

OBJECTIVE.—

To develop an evidence-based guideline to identify the optimal clinical laboratory test to identify defects in DNA mismatch repair in patients with solid tumor malignancies who are being considered for immune checkpoint inhibitor therapy.

DESIGN.—

The College of American Pathologists convened an expert panel to perform a systematic review of the literature and develop recommendations. Using the National Academy of Medicine-endorsed Grading of Recommendations Assessment, Development and Evaluation approach, the recommendations were derived from available evidence, strength of that evidence, open comment feedback, and expert panel consensus. Mismatch repair immunohistochemistry, microsatellite instability derived from both polymerase chain reaction and next-generation sequencing, and tumor mutation burden derived from large panel next-generation sequencing were within scope.

RESULTS.—

Six recommendations and 3 good practice statements were developed. More evidence and evidence of higher quality were identified for colorectal cancer and other cancers of the gastrointestinal (GI) tract than for cancers arising outside the GI tract.

CONCLUSIONS.—

An optimal assay depends on cancer type. For most cancer types outside of the GI tract and the endometrium, there was insufficient published evidence to recommend a specific clinical assay. Absent published evidence, immunohistochemistry is an acceptable approach readily available in most clinical laboratories.

Validation of a PCR-Based MGMT Promoter Methylation Assay and Comparison With Reference Lab Results

Introduction

Glioblastoma multiforme (GBM) is the most aggressive and most common primary brain tumor. Treatment of GBM and other high-grade gliomas consists of surgical resection followed by radiation and temozolomide (TMZ) chemotherapy. TMZ alkylates purine DNA bases, causing DNA damage and eventually cell death. Alkylation at O6-gauanine can be reversed by the DNA repair enzyme methylguanine methyltransferase (MGMT), inhibiting the desired cytotoxic effect. MGMT is epigenetically silenced by methylation of a CpG island in its promoter. Many high-grade gliomas are hypomethylated at the MGMT promoter. MGMT promoter methylation has emerged as a biomarker for response to alkylating chemotherapy. Multiple clinical trials demonstrate that the patients most likely to benefit from TMZ therapy have a hypermethylated MGMT promoter. MGMT promoter status is now incorporated into adjuvant therapy guidelines. Our institution receives approximately 70 to 80 high-grade glioma resections per year. The goal of this study is to validate a PCR-based MGMT promoter methylation assay and to compare results to those previously obtained at a well-known reference lab.

Methods

We extracted DNA from high-grade gliomas specimens (FFPE) or used commercially available methylated and unmethylated DNA controls. DNA was treated with bisulfite to convert unmethylated cytosine into uracil. Promoter methylation was detected with real-time quantitative PCR using a primer/probe set specific for methylated CpG sites within the MGMT promoter (meMGMT, published in Ida et al 2017) or to a housekeeping gene (COL2A1).

Results

Using control methylated DNA, we demonstrated that our assay consistently detects meMGMT and COL2A1 at DNA inputs greater than 5 ng. Both within- and between-run imprecision was minimal with little variation in crossing thresholds (Cts) for both COL2A1 (%CV at 50 ng input: 7.9% [within-run], 0.3% [between-run]) and MGMT (%CV at 50 ng input: 0.6% [within-run], 0.4% [between-run]). Delta Ct values were consistent across different levels of DNA input. Analysis of 15 cases from our archives (8 positive, 7 negative) demonstrates that our results (delta Ct values) correlate (R2 = 0.9) with the reference lab methylation score (<2 considered unmethylated). Separation between positive and negative cases was good with minimal overlap; positive cases had delta Cts of 0.2 to 7.6, while negative cases had delta Cts of 7.1 to 13.6 or had no meMGMT amplification. ROC curve analysis showed an AUC of 0.96, and using the cutoff of delta Ct <7.6, our assay had 93% (14/15) agreement with the reference lab, a sensitivity of 100%, and a specificity of 71%. The single negative case that was misclassified had a reference lab methylation score of 1.86, near their cutoff of <2. Additional patient samples are pending.

Conclusions

Our PCR-based MGMT methylation assay has excellent analytical sensitivity and low imprecision. Preliminary analysis of patient samples shows that our results are concordant with historic reference lab results.

Clonal repopulation dynamics in recurrent human papillomavirus-associated head and neck cancer

e17517

Background: Despite the better prognosis for human papillomavirus (HPV)-associated head and neck cancer (HNC) relative to HPV-negative HNC, 10 to 25% of HPV-associated cases recur within 3 years of completing therapy. We sought to investigate clonal repopulation dynamics in recurrent HPV-related oropharyngeal squamous cell carcinoma (OPSCC). Methods: Deep sequencing ( > 500X) using a CLIA-certified, targeted 262-gene multiplexed assay was performed on DNA extracted from 11 pairs of matched FFPE-preserved primary and subsequent recurrent p16-positive OPSCC samples including two pairs from the University of Washington and nine from the University of Pittsburgh. Results: 82% of patients were male with a median (IQR) age of 62 (15) years. Nine of eleven patients presented with advanced stage disease. Five were initially treated with surgery and concurrent chemoradiation, two with surgery, and four with concurrent chemoradiation. Median (IQR) time to recurrence was 10 (10.5) months. Four patients reported a history of tobacco use. The mean (SD) number of somatic nucleotide variants (SNVs) per tumor was 16.4 (12.2). 42% of SNVs occurred in the primary alone while 28% were unique to recurrences. 28% of SNVs were shared between primary-recurrent pairs. Three patients with distant or regional recurrences had different mutational profiles between the primary and recurrence suggesting repopulation by an early-appearing clone vs sampling error. One patient developed two local recurrences with conserved mutations in addition to de novo mutations in each subsequent recurrence suggesting repopulation by early and late evolving clones. Genes with a substantial functional impact that were frequently mutated in primary tumors include IDH2 and FBXW7. Recurrently mutated genes found in recurrent tumors include TET3, PIK3CA, STK11, and HDAC4. Conclusions: In thistargeted exome sequencing study of matched primary-recurrent OPSCC tumor pairs, we observed clonal repopulation with both transmitted and de novo mutations. Intriguingly, we noted a predilection for mutations involving genes regulating metabolism including IDH2 among primary tumors which has potential implications for precision medicine approaches.

Targeted Next-Generation Sequencing of Acute Leukemia

Mutation profiling of acute leukemias is a valuable tool for identifying genetic mutations with prognostic, predictive, therapeutic, and diagnostic utility. Technological advances, such as massively parallel sequencing, have allowed laboratories to assess for variation across dozens or hundreds of genes simultaneously with relatively low cost per target.Here, we describe a procedure for designing and using a TruSeq Custom Amplicon assay targeting genes involved in acute leukemias. This method is a fully customizable, amplicon-based assay for targeted resequencing, allowing interrogation of selected genomic regions of interest. The most readily available form of the assay allows sequencing of up to 1536 amplicons in a single reaction using a straightforward workflow. The ability to multiplex up to 1536 amplicons per reaction allows coverage of up to 650 kb of cumulative sequence and supports up to 96 samples per batch, depending on library size and desired sequencing depth.