Genomic alterations involving the CRLF2 gene lead to over-expression of intact CRLF2 and have significant prognostic value in pediatric BCP-ALL. Not only do patients with these lesions have inferior outcomes, they also have a very high frequency of JAK1 and JAK2 mutations and may be candidates for targeted therapies. The two major CRLF2 lesions include cryptic translocations that produce IgH@-CRLF2 and interstitial deletions of the pseudoautosomal region of X/Y causing P2RY8-CRLF2 fusion. Both lesions can be detected by fluorescence in situ hybridization (FISH), and genomic PCR or RT-PCR can identify P2RY8-CRLF2. To develop rapid and inexpensive assays for detection/screening of these events, we developed a flow cytometry based method to measure CRLF2 expression and compared this assay to quantitative RT-PCR (qPCR) measurement of CRLF2 expression by evaluating their performance in an unselected cohort of 279 newly diagnosed pediatric BCP-ALL patients consecutively enrolled on the COG AALL03B1 biology/classification study between 10/30/09-5/1/10. Flow cytometry was performed first in real time on diagnostic specimens shipped to a central COG reference laboratory and then residual diagnostic material was shipped to a separate laboratory for RNA isolation and qPCR analysis. Of the 279 cases analyzed by flow, 257 (92%) yielded sufficient RNA quality and quantity for qPCR analysis. In our previous studies with qPCR and CRLF2 it was shown that CRLF2 lesions occurred only among those cases with the highest expression (ΔCt < 8). In order to assure that we identified all cases with CRLF2 lesions, we performed FISH and P2RY8-CRLF2 PCR on all cases with qPCR expression ΔCt < 10 (n = 109) and an additional 14 cases with a flow blast/lymph CRLF2 mean fluorescence intensity (MFI) ratio >1.15. Of these 123 cases, 11 were determined by FISH to have the IGH@-CRLF2 translocation and 15 were shown to have P2RY8-CRLF2 fusions by PCR. Figure 1 shows the locations of these genomic lesion-positive cases among the qPCR (panel A) and flow cytometry (panel B) CRLF2 expression data. The overall frequency of CRLF2 lesions among these patients is 10.1% (assuming all lesions were identified among the highest expressing cases) and, surprisingly, the frequencies of IgH@-CRLF2 and P2RY8-CRLF2 were very similar (4.3% and 5.8%, respectively). With both methods, the 11 IgH@-CRLF2 cases were found to be the highest expressing (among the top 12 cases by qPCR and 16 cases by flow). Receiver operating curve analysis of each method identified cutoffs with excellent performance: qPCR cutoff CRLF2 ΔCt = 5.47 with 96.9% specificity and 88.5% sensitivity; flow cutoff MFI CRLF2 ratio of 2.04 with 95.9% specificity and 92.3% sensitivity. The broader dynamic range of the qPCR assay may be necessary for the identification of poor risk cases with high CRLF2 expression that lack genomic lesions, however both methods are rapid, highly effective and very comparable for finding ALL cases that harbor CRLF2 genomic lesions, and suitable for incorporation in large scale clinical trials.
No relevant conflicts of interest to declare.
This icon denotes a clinically relevant abstract
Asterisk with author names denotes non-ASH members.