Development of a DNA-Based Targeted Assay Suitable for Accurate Quantification of Chromosomal Deletions in Myelodysplastic Syndromes with Deletion (5q) and Other Clonal Diseases

Introduction

The acquisition of large-scale chromosomal lesions is a frequent event in malignant disorders. One example is the recurrent deletion of chromosome 5q (del(5q)) in myelodysplastic syndromes (MDS). The detection and monitoring of such deletions are important elements in routine clinical diagnostics and cancer genomic studies. Currently established methods for their assessment are metaphase cytogenetics (MC), fluorescence in situ hybridization (FISH) and micro-array based techniques. However, each of these methods harbours specific disadvantages as they depend on (viable) cells, are expensive, labour-intensive or only semi-quantitative. One possible approach to interrogate chromosomal deletions constitutes the assessment of allelic loss at heterozygous short tandem repeat (STR) loci within deleted regions. Therefore, we aimed to establish a robust, quick and inexpensive PCR assay that measures allelic imbalance at such STR loci in order to reliably estimate frequencies of cells carrying del(5q) from only minute amounts of DNA.

Methods

Genomic DNA (gDNA) was isolated from bone marrow (BM) or blood cells of MDS and acute myeloid leukemia (AML) patients with cytogenetically confirmed del(5q). Based on NCBI UniSTS database, we designed 12 fluorochrome-labelled PCR amplicons with size ranges of 100-400 bp that surround STR loci between chromosomal bands 5q21 and 5q31. Using only 10 ng gDNA, all 12 PCR amplicons were amplified in a single optimized multiplex-PCR reaction. Subsequently, amplicon fragment analysis was carried out via capillary electrophoresis on an ABI 3130 Genetic Analyzer. Allele size quantification of informative heterozygous loci was performed using ABI Genemapper software. Furthermore, size calculations of individual alleles were corrected for PCR-stutter, which was estimated from corresponding loci in homozygous samples. Finally, using mesenchymal stromal cells as a germline control, the degree of skewing in the allelic ratios of all informative STR markers in the tumor sample relative to the corresponding allelic ratios in the control was averaged and subsequently translated into fractions of cells carrying the del(5q) lesion.

Results

Application of our novel assay for quantification of del(5q) burden in n=559 samples from n=67 patients revealed a high frequency of informative markers with an average of 7 heterozygous STR loci per patient. The data shows a strong inter-marker concordance with a standard deviation (SD) of 2.3% for del(5q) cell frequencies. Moreover, duplicate analysis of 328 samples revealed an average SD of 0.86%. Most importantly, paired analysis of the proportion of del(5q) cells estimated using interphase-FISH and our PCR assay was carried out for n=9 samples and resulted in strong correlation with r²=0.93. A serial dilution series with deleted and non-deleted gDNA also revealed highly concordant results with r²=0.96. When comparing matched germline and tumor STR profiles, no case of microsatellite instability was detectable in our MDS/AML cohort thus highlighting the suitability of STR based lesion quantification in this disease entity. Furthermore, we could use our large dataset to calculate amplification efficiencies for each locus in order to predict “surrogate” germline profiles eventually allowing us to calculate del(5q) frequencies in tumor samples that lack germline controls. Finally, our assay reliably confirmed molecular remission in BM from del(5q) MDS patients after Lenalidomide treatment, in agreement with concomitant MC analyses.

Conclusion

Using a multiplexed PCR assay for measurement of STRs in deleted chromosomal regions we present a highly adaptable tool for precise quantification of large scale lesions. We show a very good correlation with established methods exemplarily for del(5q) lesions. Even without available germline control our assay provides robust results. Requiring only minute amounts of gDNA, this assay is ideally suitable for copy number quantification in samples for which only residual archival gDNA and no cells for FISH analysis are available. Due to the small amplicon sizes it should also be useful for investigation of fragmented gDNA from formalin-fixed archival specimen. In summary, our newly developed assay offers a mean to obtain quantitative data for basically any large scale chromosomal deletion which contains STRs and should be easily applicable to other clonal diseases.