Quantification of within-sample genetic heterogeneity from SNP-array data.

Area of Science:

• Genomics
• Cancer Biology
• Bioinformatics

Background:

• Intra-tumour genetic heterogeneity (ITH) is a major challenge in cancer treatment, often leading to drug resistance.
• Current methods for measuring ITH, like multi-region sampling, are expensive and difficult to implement routinely.
• There is a need for accessible tools to estimate ITH from standard genomic data, such as single nucleotide polymorphism arrays (SNP-arrays).

Purpose of the Study:

• To develop and validate novel computational scores for quantifying ITH from single SNP-array samples.
• To assess the accuracy of these scores in estimating clonal diversity using in-silico and in-vitro models.
• To evaluate the prognostic potential of the developed ITH scores in predicting patient survival.

Main Methods:

• Designed two new scores, S (Shannon diversity index) and R (Ripley's L statistic), to quantify ITH in individual SNP-array samples.
• Utilized in-silico and in-vitro mixtures of tumour clones for benchmarking the developed scores.
• Correlated the novel scores with existing ITH estimates derived from sequencing data and assessed their prognostic value.

Main Results:

• The novel scores showed significant but variable associations with known diversity in in-silico models (p < 0.001).
• Moderate associations with known diversity were observed in in-vitro models (p = 0.015 and p = 0.085).
• The scores demonstrated moderate prognostic potential, significantly predicting survival across several tumour types (corrected p = 0.03), although tumour cell fraction heterogeneity impacted quantification accuracy.

Conclusions:

• Individual SNP-arrays can provide insights into intra-sample clonal diversity with moderate accuracy.
• The developed scores offer a potential method for routine ITH estimation using standard genomic data.
• Further refinement is needed to overcome limitations like tumour cell fraction variability for more precise ITH quantification.