Cancer Diagnosis Program | Biorepositories and Biospecimen Research Branch

Search BRD Papers

Method of Tissue Acquisition Affects Success of Comprehensive Genomic Profiling in Lung Cancer.

Author(s): Mata DA, Harries L, Williams EA, Hiemenz MC, Decker B, Tse JY, Janovitz T, Ferguson DC, Speece IA, Margolis ML, Mathews B, Fedorchak K, Killian JK, Xiao J, Tolba KA, Ramkissoon S, Vergilio JA, Elvin JA, Oxnard GR, Ross JS, Huang RSP

Publication: Arch Pathol Lab Med, 2022, Vol. , Page

PubMed ID: 35771716 PubMed Review Paper? No

Purpose of Paper

Conclusion of Paper

Studies

1. Study Purpose

   This study retrospectively examined data from lung carcinoma specimens submitted for comprehensive genomic profiling (CGP) to identify potential effects of specimen acquisition method, specimen size, specimen cellularity, specimen anatomical location, specimen submission as slide versus FFPE block, and patient age, sex, and diagnosis on the CGP success rate. This study included data from 3,312 specimens from lung carcinoma patients (median age 69 years, 1,699 males and 1,613 females) that were submitted for comprehensive genomic profiling (CGP) over a two month period. Specimens included 321 FFPE lung resections; 2,236 FFPE lung biopsies; 149 FFPE bone biopsies; 174 FFPE cell blocks from pleural, peritoneal, or pericardial fluid; and 432 FFPE cell blocks from fine needle aspirations (FNAs), but details of specimen acquisition and processing were not described.  All diagnoses were confirmed by examination of H&E stained slide by a pathologist. In total, 3,191 specimens were non-small cell lung carcinomas (NSCLCs) and 121 small cell lung carcinomas (SCLCs). Of the 3,191 NSCLCs, 2,675 were subtyped and these included 2,054 adenocarcinomas, 546 squamous cell carcinomas, 35 large cell neuroendocrine carcinomas, 22 adenosquamous carcinomas, and 18 sarcomatoid carcinomas. Sections were macrodissected to ensure >20% tumor nuclei cellularity (TNC). DNA was extracted from unstained FPFE slides and cell blocks using an unspecified method. A minimum of 55 ng of DNA was used for 500X next generation sequencing of the exons of 309 cancer related genes and the introns of 34 genes. Specimens were considered to have passed pre-sequencing quality control if they had ≥20% estimated TNC, ≥25 mm² tissue surface per slide or ≥1.0 mm³ tissue volume, and ≥55 ng DNA was isolated; however, sequencing was also conducted at the physician’s discretion for specimens with 10-20% TNC, 5-25 mm² tissue surface area or with a volume 0.6-1.0 mm³, and those yielding 27-55 ng DNA. Cases were considered successful if sequencing achieved ≥250X median coverage, ≥100X coverage at 95% of sites, <1% contamination and ≥25% computational purity.  In cases where a pathogenic alteration was identified but one sequencing quality control metric was not met a qualified report was made.

   Summary of Findings:

   The median TNC was 20%, with 78.8% of specimens having the recommended estimated TNC ≥20%, 8.6% of specimens had a TNC of 10-20% (CGP at the physician’s discretion), and 12.3% had a TNC of <10% (excluded from sequencing). In total, 89.3% (2,956) of specimens underwent DNA extraction, of those 203 failed to yield sufficient DNA. Overall, 2,321 of the 3,312 specimens were successful (all sequencing quality control parameter met) in CGP and for another 510 a qualified report was generated (failed one sequencing quality control metric, but a pathogenic alteration was identified). CGP success was significantly affected by specimen type (P<0.001) and significance was maintained after multivariable adjustment for patient age, patient sex, diagnosis, specimen location, TNC, tissue surface area, and submission of block versus slides.  The CGP failure rate was only 2.8% (9 of 321) for surgical excision specimens, but 14.5% (324 of 2236) for biopsy specimens, 18.5% (80 of 432) for FNA specimens, 18.8% (28 of 149) for bone biopsy specimens and 23% (40 of 174) for fluid cytology specimens.  CGP failure rate was higher for non-neuroendocrine NSCLC specimens than small cell or large cell lung neuroendocrine carcinoma specimens (14.8% versus 9.1% and 5.7% respectively; P =0.001). CGP failure rate was slightly higher for primary (biopsy of resection) than metastatic (biopsy or resection) tumor specimens (16.2% versus 12.6%, P<0.001).  Among metastases, brain lesions had the lowest failure rate (1.7%, 2 of 120) and bone had the highest (18.8%, 28 of 149). Notably, 16 of the 28 bone metastases that failed to sequencing were due to failure to obtain extract sufficient DNA. CGP failures were higher for specimens with <20% visually estimated TNC values versus the recommended ≥20% (P<0.001) or specimens below the recommended size (<25 mm²) than those ≥25 mm² (P<0.001). While there was a trend toward increased failure in specimens submitted as unstained slides versus FFPE blocks, the difference was not significant. No effect of patient age, or gender on CGP success was found. The multivariable model confirmed a significant association between specimen site (metastasis versus primary tumor), specimen format (slide versus FFPE block), tissue surface area and TNC on CGP failure (P<0.001, all), but found no association of CGP failure with patient age, sex or diagnosis.  Of the 2,891 specimens suitable for sequencing, 2,834 (98%) had at least one reportable pathogenic genomic alteration. Further analysis found one or more single nucleotide variation(s) or indel(s) in 2,761 cases (95.5%), a focal copy number variation in 1,742 cases (60.3% of cases), and a rearrangement or fusion in 514 cases (17.8%). Of these cases, 42.2% had at least one targetable genomic alteration identified

   Biospecimens

   • Tissue - Bone
   • Tissue - Lung
   • Cell - Lung
   • Bodily Fluid - Pericardial Fluid
   • Bodily Fluid - Pleural Fluid
   • Bodily Fluid - Peritoneal Fluid

   Preservative Types

   • Formalin

   Diagnoses:

   • Neoplastic - Carcinoma

   Platform:

   Analyte	Technology Platform
   DNA	Next generation sequencing

   Pre-analytical Factors:

   Classification	Pre-analytical Factor	Value(s)
   Preaquisition	Patient age	62-77 Years
   Preaquisition	Patient gender	Female Male
   Storage	Storage conditions	As FFPE block As unstained section
   Biospecimen Aliquots and Components	Aliquot size/volume	≥25 mm2 tissue surface per slide or ≥1.0 mm3 tissue volume 5-25 mm2 tissue surface per slide or 06-1.0 mm3 tissue volume <5-25 mm2 tissue surface per slide or <06 mm3 tissue volume
   Biospecimen Aliquots and Components	Biospecimen heterogeneity	Differing levels of cellularity
   Biospecimen Aliquots and Components	Biospecimen components	<10% TNC 10-20% TNC ≥20% TNC
   Biospecimen Acquisition	Method of tissue acquisition	Surgical resection Fine needle aspiration Biopsy
   Preaquisition	Diagnosis/ patient condition	Non-small cell lung carcinomas Small cell lung carcinoma Adenocarcinoma Squamous cell carcinoma Large cell neuroendocrine carcinoma Adenosquamous carcinoma Sarcomatoid carcinoma
   Biospecimen Acquisition	Biospecimen location	Lung Bone Pleural, peritoneal, or pericardial fluid Primary tumor Metastasis

   View more