Capture Sequencing Is a Useful Method for Comprehensive Clonality Analysis Based on Ig/TCR Gene Rearrangements in Acute Lymphoblastic Leukemia

Introduction

Immunoglobulin (Ig)/ T-cell receptor (TCR) gene rearrangements are the most widely used clonal marker to detect residual leukemic cells in patients with Acute Lymphoblastic Leukemia (ALL). Ig/TCR gene rearrangements based molecular minimum residual disease (MRD) monitoring has become one of the most powerful prognostic indicators for patients with ALL. Although the standard method of real-time quantitative PCR (RQ-PCR) provides very good sensitivity in MRD measurement, the workflow is very complicated and time-consuming, requiring expert technique and much work for operation, which limits the number of patients to be examined for MRD monitoring and the number of testable markers per patient.

Material and methods

To reveal clonal architecture and detect appropriate MRD marker, we designed capture probes covering the coding and recognition signal sequences of V, D, J genes of the Ig/TCR loci. We performed high-throughput target-capture sequencing in 208 pediatric cases with BCP-ALL and 35 pediatric cases with T-cell ALL, including 20 relapsed cases and 14 MRD marker negative cases. Extracted DNA samples were enriched with about 420 capture probes (Agilent Technology) and sequenced by HiSeq2500 platform (Illumina) in order to obtain enough sequence coverage (> 500 mean depth). Sequenced data were analyzed with Ig/TCR recombination analysis tool Vidjil (Giraud et al, 2014) and V(D)J recombination clones were listed according to a number of detected read for each clone.

Results

Total 2379 clonal Ig/TCR gene rearrangements (median 9 per patient, range 0-82) were detected by capture sequencing among 236 (97%) cases. A clonal IGH sequence with V(D)J recombination was identified in 91% of BCP-ALL cases, followed by TRG (68 %), IGK (67 %), TRA+D (66%), TRD (59 %), TRB (49%), and IGL (15 %), respectively. On the other hand, clonal TRG V(D)J recombination was detected in 74% of T-ALL cases, followed by TRB (69%), TRD (57%), IGH (26%), and TRA+D (6%), respectively. About half of BCP-ALL cases were identified two independent IGH rearrangements. These frequencies agree with previous reports obtained by PCR based experiments.

In the cases in this study with well-characterized clonal Ig/TCR gene rearrangements by PCR and Sanger sequencing, our capture sequencing was able to detect all rearrangements used in MRD measurements. Although 8 BCP-ALL cases in this study were marker-negative in standard PCR-MRD diagnostics, clonal Ig/TCR gene rearrangements were identified for 5 out of 8 cases by capture sequencing. Some of the hidden clonal rearrangements showed specific and good quantitative amplification by RQ-PCR and can be used as sensitive PCR-MRD targets. On the other hand, all the MRD marker negative 6 T-ALL cases were not detected clonal Ig/TCR gene rearrangements.

Finally, we compared the clonal architecture based on Ig/TCR gene rearrangements between diagnosis and relapse in relapsed B-ALL patients. Changes in the clonal architecture were associated with remission duration. In very early relapse cases, detected Ig/TCR rearrangements and their proportion at relapse are very similar to those at diagnosis. In early to late relapse cases, some major Ig/TCR gene rearrangements were lost at relapse and other minor rearrangements expanded at relapse. Most of the identified Ig/TCR gene rearrangements were different between at diagnosis and at relapse in a case relapsed after more than 10 years. Loss of rearrangements were commonly seen in TRA, TRB, and IgL, while most of the IgK and TRD rearrangements were steady during disease course.

Conclusion

Introducing target capture sequencing enables to high throughput sample preparation and automated data analysis. Capture sequencing is a useful method for comprehensive detection of Ig/TCR gene rearrangements and contributes to better understanding clonal architecture and detecting appropriate MRD markers in ALL patients.