Assessing the Feasibility and Limitations of Cultured Skin Fibroblasts for Germline Genetic Testing in Hematologic Disorders

Introduction

Peripheral blood is the standard tissue source for germline genetic testing in most scenarios. In patients with hematologic malignancies, however, peripheral blood frequently contains tumor- or clonal hematopoiesis-related acquired genetic variants, often occurring in genes that can also cause inherited cancer susceptibility if present in the germline. Thus, an alternative tissue source is necessary. Cultured skin fibroblasts have been used as a potentially ideal alternative because they are free from blood contamination and provide ample DNA yields, advantages that other alternatives such as saliva or nail clippings lack. However, optimal culture methods, expected time from biopsy to sufficient DNA yield, culture failure rate, and limitations of this technique, including the possibility of variants being acquired solely due to the culturing process, are not yet known.

Methods

We conducted a retrospective cohort study of subjects with cytopenias or hematologic malignancies who underwent skin biopsy and fibroblast culture for germline genetic testing from April 2014 to June 2018. Skin biopsy culture technical data, including time from biopsy to culture set-up, shipment from an outside institution, culture failure, and biopsy size, were abstracted from tissue culture logs. Patient demographics, comorbidities, medication history, and hematologic diagnosis and treatment were abstracted from medical records. Next generation sequencing data from targeted capture of 144 inherited cancer and bone marrow failure predisposition genes obtained for clinical genetic testing purposes were analyzed to identify variants at both germline (40-60%) and subclonal (10-40%) variant allele frequencies (VAF). Pathogenicity was interpreted according to ACMG/AMP guidelines. Fisher's exact tests and logistic regression models were used to assess associations with culture failure. T-tests and linear regression models were used to assess factors associated with mean time to confluency.

Results

In total, we studied 350 samples from unique patients, including 61 (24%) who carried one or more pathogenic or likely pathogenic cancer susceptibility gene variant(s). Overall, 16 of the 350 (5%) biopsies failed to grow in culture. The median time from skin biopsy to sufficient growth to extract DNA for genetic testing was 27 days (IQR 22-29 days). Culture failure was significantly more likely in samples with a delay in culture initiation for 24 hours post biopsy (OR=4.32; p<0.01), and a pathogenic germline variant in a gene associated with telomere maintenance (OR=64.50; p<0.01). Factors associated with an increased mean time to sufficient growth included prior allogeneic stem cell transplant (32.1 days versus 27.2 days; p<0.01) and prior intravenous (IV) steroid exposure (29.9 days versus 26.4 days; p<0.01). Among samples cultured successfully, carriers of any pathogenic germline variant had a significantly decreased mean time to sufficient growth (25.4 days versus 28.6 days; p<0.01). A pathogenic or likely pathogenic subclonal variant was identified in 11 (4%) subjects at a median VAF of 20%. Among eight of these with additional tissue available, the presence of the variant was confirmed in four (50%). In individual cases, we found evidence of loss of a pathogenic variant in the hematopoietic malignancy. In one patient with a pathogenic variant with a 50% VAF in the original skin culture, the variant was not present in a skin culture from a second, fresh skin biopsy done due to discordant phenotype.

Conclusions

Culturing of skin fibroblasts for germline genetic testing in patients with hematologic disorders has a high success rate, especially when cultures are initiated within 24 hours of collection, and adds on average 27 days to genetic testing turnaround time. From patients with a hereditary syndrome, most skin biopsies will culture with the exception of individuals with a short telomere syndrome. For this subset, a direct skin biopsy without culture may be necessary. Subclonal variants at VAFs relevant to interpretation of a germline test were found in 4% of cases. Half were confirmed in an alternative tissue. Etiology of the subclonal variants, whether acquired during the culturing process or due to mosaicism or sequencing biases was not always clear. Careful assessment of the clinical phenotype in interpreting and applying germline genetic results to patient care will always be warranted.