Cancer Diagnosis Program | Biorepositories and Biospecimen Research Branch

Search BRD Papers

Association Between Preanalytical Factors and Tumor Mutational Burden Estimated by Next-Generation Sequencing-Based Multiplex Gene Panel Assay.

Author(s): Quy PN, Kanai M, Fukuyama K, Kou T, Kondo T, Yamamoto Y, Matsubara J, Hiroshima A, Mochizuki H, Sakuma T, Kamada M, Nakatsui M, Eso Y, Seno H, Masui T, Takaori K, Minamiguchi S, Matsumoto S, Muto M

Publication: Oncologist, 2019, Vol. , Page

PubMed ID: 31186376 PubMed Review Paper? No

Purpose of Paper

Conclusion of Paper

Studies

1. Study Purpose

   This study retrospectively investigated the effects of DNA library concentration, formalin-fixation, storage duration, sampling primary versus metastatic tumor tissue, sampling within radiation field, and number of chemotherapy regimens on the TMB in unmatched specimens. Sequencing data generated from 217 solid tumor specimens (36 pancreas, 25 colon, 24 biliary tract, 18 unknown, 14 stomach, 13 ovarian, 12 esophagus, 9 urinary, 9 uterine/cervical, 8 lung, 7 breast, 6 neuroendocrine, 6 liver, 6 head and neck, 6 bone, and 4 unspecified tumor specimens) between April 2015 and April 2018 was analyzed. The specimens were selected for sequencing based on being rare tumors, cancers of unknown primary site, and cancer refractory to standard chemotherapy. As the study analyzed the data retrospectively, no details of specimen procurement or processing were provided but 158 of the 217 specimens were FFPE sections and the remaining were frozen. Of the 217 specimens, 199 were stored less than 5 years. NGS data was generated using the OnePrime NGS gene panel assay which includes 215 genes and 17 selected regions. The median depth of sequencing was 3,000. Variants were identified with VarProwl using a standard of identification on both strands at similar proportion (bias <0.30) and the probability of encountering a variant at that location is much higher (P<0.001) than the error rate at that position for a VAF of 30. To determine the TMB, the number of variants that occurred in the specimen at 4-95% but do not occur in >1% of the population were counted and divided by the size of the coding region of the included genes (0.78 MB). Specimens were classified as low (<10 mutations/MB) or high (≥10 mutation/MB). Tumor libraries were quantified using the KAPA library quantification kit. The effect of storage was examined in the 199 specimens stored less than 5 years. The effect of radiation was investigated in esophageal cancer specimens with library yields ≥5 nM by comparing TMB in five specimens outside the radiation field and three within the radiation field. The effect of chemotherapy regimens was investigated by comparing the TMB in specimens obtained in patients after 0 (118 patients), 1 (39 patients), 2 (23 patients), or ≥3 (19 patients) chemotherapy regimens.

   Summary of Findings:

   Overall, the median TMB was 10.4 mutations per MB and the median variant allele frequency (VAF) was 0.0642 but 52% of specimens had high TMB (≥10 mutations/MB) and the median VAF for these specimens was 0.0515. The TMB was significantly higher in specimens with a library DNA concentration of <5 nM than those with a concentration ≥5 nM (P<0.01) and all 50 cases with more than 50 mutations/MB had library concentrations of <5 nM. As expected, the mean library concentration was higher for specimens with a TMB of ≤50 mutations/MB (P<0.001). The median TMB score was significantly higher in FFPE specimens than frozen specimens when all specimens were considered (11.7 versus 5.2 mutations per MB, P<0.01) and when just those with a library concentration of ≥5 nM were considered (9.1 versus 5.8 mutations per MB, p<0.05). TMB was comparable in primary and metastatic sites (11.7 and 9.1 mutations per MB, respectively). There was a weak correlation between the storage duration (date of procurement to date of NGS) and the TMB for specimens stored less than 5 years (r=0.39, P<0.001), but the significance disappeared if only specimens with a library concentration of ≥5 nM were considered. The median TMB was higher in the three esophageal specimens obtained from lesions within the radiation field than those from the non-radiation field (15.6 versus 5.2 mutations per MB, P=0.02). Interestingly, the TMB was similar in specimens subjected to 0, 1, 2, and 3 or more chemotherapy regimens and was not high (5.19 mutations per MB) in the one patient with a history of temozolomide.

   Biospecimens

   • Tissue - Bone
   • Tissue - Pancreas
   • Tissue - Not specified
   • Tissue - Head and Neck
   • Tissue - Colorectal
   • Tissue - Esophagus
   • Tissue - Ovary
   • Tissue - Stomach
   • Tissue - Breast
   • Tissue - Urine
   • Tissue - Uterus
   • Tissue - Liver
   • Tissue - Nerve

   Preservative Types

   • Frozen
   • Formalin

   Diagnoses:

   • Neoplastic - Not specified

   Platform:

   Analyte	Technology Platform
   DNA	Next generation sequencing

   Pre-analytical Factors:

   Classification	Pre-analytical Factor	Value(s)
   Biospecimen Acquisition	Biospecimen location	Primary tumor Metastasis Within radiation field Not within radiation field
   Preaquisition	Other drugs	No chemotherapy One chemotherapy regimen Two chemotherapy regimens Three or more chemotherapy regimens
   Biospecimen Preservation	Type of fixation/preservation	Formalin (buffered) Frozen
   Storage	Storage duration	0-60 months
   Next generation sequencing Specific	Template/input amount	Library DNA concentration <5 nM Library DNA concentration ≥5 nM

   View more