Specific and Sensitive Diagnosis of BCOR-ITD in Various Cancers by Digital PCR

Abstract

BCOR is an epigenetic regulator altered by various mechanisms including BCOR-internal tandem duplication (BCOR-ITD) in a wide range of cancers. Six different BCOR-ITD in the 3'-part of the coding sequence of exon 15 have been reported ranging from 89 to 114 bp in length. BCOR-ITD is a common genetic alteration found in clear cell sarcoma of the kidney and primitive myxoid mesenchymal tumor of infancy (PMMTI) and it characterizes a new type of central nervous system tumor: "CNS tumor with BCOR-ITD". It can also be detected in undifferentiated round cell sarcoma (URCS) and in high-grade endometrial stromal sarcoma (HGESS). Therefore, it is of utmost importance to search for this genetic alteration in these cancers with the most frequent technique being RNA-sequencing. Here, we developed a new droplet PCR assay (dPCR) to detect the six sequences characterizing BCOR-ITD. To achieve this goal, we used a single colored probe to detect both the duplicated region and the normal sequence that acts as a reference. We first generated seven synthetic DNA sequences: ITD0 (the normal sequence) and ITD1 to ITD6 (the duplicated sequences described in the literature) and then we set up the optima dPCR conditions. We validated our assay on 19 samples from a representative panel of human tumors (9 HGNET-BCOR, 5 URCS, 3 HGESS, and 2 PMMTI) in which BCOR-ITD status was known using at least one other method including RNA sequencing, RT-PCR or DNA-methylation profiling for CNS tumors. Our results showed that our technique was 100% sensitive and specific. DPCR detected BCOR-ITD in 13/19 of the cases; in the remaining 6 cases additional RNA-sequencing revealed BCOR gene fusions. To conclude, in the era of histomolecular classification of human tumors, our modified dPCR assay is of particular interest to detect BCOR-ITD since it is a robust and less expensive test that can be applied to a broad spectrum of cancers that share this alteration.

Keywords: BCOR-internal tandem duplication; FFPE tissue; HGNET-BCOR; diagnostic marker; digital PCR assay.

Figure 1

(A) dPCR for tandem duplication in one channel: W1 no template control; W2 normal sample; W3 homozygous ITD BCOR; W4 heterozygous ITD BCOR; forward primer; reverse primer; probe. (B) Position of primers and probe on synthetic sequences. ITD0: normal sequence; ITD1 to ITD6 sequences with internal tandem duplication (ITD). Forward primer in cyan; reverse primer in green; probe in yellow; invariable duplicated sequence in upper case; A: amplicon size; classic amplicon is underlined; long amplicon is underlined with dashes.

Figure 2

Validation of the optima dPCR conditions. Three droplet clusters differ according to their fluorescence levels: lowest signal (1,000< >3,000 AU) for negative droplets (below black line), “normal” cluster (3,000< >5,000 AU) have the same intensity level as ITD0_p (between black and pink line) and a third cluster of higher intensity (>5,000 AU) corresponding to a characteristic “BCOR-ITD” signal (above pink line). (A1) Nanodroplet cluster intensity for each pure artificial ITDs_p . (A2) Log of the number of events as a function of the amplitude of the fluorescence allows a single threshold to be set for all samples and isoforms (pink line). (B1) Nanodroplet cluster intensity for each six mixed ITDs₅₀ with 50% of ITD0_p. (B2) Log of the number of events as a function of the amplitude of the fluorescence allows a single threshold to be set for all samples and 50% mixed isoforms (pink line). C+, FFPE positive control; NTC, no template control; ITDs_p, pure synthetic DNA sequences; ITDs₅₀, 50% mixed ITD1_p to ITD6_p with ITD0_p.

Figure 3

Sensitivity of BCOR-ITD detection. Results for serial dilutions of ITD1_R (A1, 2, 3) and ITD4_R (B1, 2, 3). (A1, B1) Linear correlation between expected fractional abundance of ITDs (x) and measured values (y). (A2, B2) The 1D plot showing the number of events for each fluorescence amplitude allowing to set a single threshold for all samples and isoforms (pink line). (A3, B3) The 1D plot showing for each dilution the positive events (BCOR-ITD) corresponding to the data points above the pink baseline marker, negative events at the bottom of the plot and reference events (BCOR-wildtype) between the two. C+, FFPE positive control; FA, fractional abundance.

Figure 4

Example of BCOR-ITD detection by dPCR assay for six samples (BCOR-1, BCOR-17, BCOR-4, BCOR-6, BCOR-16, and BCOR-3). The 1D plot shows for each sample the positive events (BCOR-ITD) corresponding to the data points above the pink baseline marker, negative events at the bottom of the plot and reference events (BCOR-wildtype) between the two. Internal controls are analyzed at the same time: SW48 and H1650 cell lines which are BCOR-wildtype, duplicated and non-duplicated cases (C+ and C−), as well as NTC (no template control = water).

Figure 5

The place of dPCR for BCOR-ITD in the diagnostic workflow to reach a final diagnosis.