Abstract
Introduction: Chronic lymphocytic leukemia is a clonal proliferation and accumulation of neoplastic small B cells in the peripheral blood, bone marrow and lymph nodes. Many patients with CLL relapse even after aggressive therapies. Several studies have suggested that residual leukemic cells are responsible for relapse. Hence, it is important to have a highly sensitive assay to detect Minimal Residual Disease (MRD) in patients during treatment or when the patient is in morphologic remission. Assays that test for patient specific immunoglobulin heavy chain (IGH) gene rearrangement should permit higher sensitivity for MRD detection. We developed a real-time PCR based MRD assay (LymphoTrack™ assay) that has a minimum limit of detection of one tumor cell in 10,000 normal cells, which is more sensitive than FACS based methods.
Methods: Clonal, patient-specific IGH DNA sequences were obtained from 7 patient samples. Multiple DNA samples from normal human tonsils were used as normal polyclonal controls for all studies. Primer sets were designed using an upstream patient-specific primer paired with either a reverse allele-specific primer targeting the JH intron or a second patient-specific primer. In either of the methods, the primer pair was combined with an allele specific TaqMan probe targeted at an appropriate region of the IGH sequence. For any given patient sequence, multiple primer sets were designed, and then run on a series of specificity and sensitivity tests to select the best primer set. First to test for specificity, all primer sets were run with various lots of normal tonsil DNA and water using SYBR green detection. The primer sets that didn’t amplify tonsil DNA or had an amplification product with a Ct > 40 were further tested for specificity with the same tonsil DNAs and water by including the appropriate TaqMan probe for that patient. The primer/probe sets that did not amplify either tonsil or water were then tested with the patient DNA. Diagnostic patient DNA was serially diluted into tonsil DNA to verify the ability of the primer/probe set to identify at least a 10−4 dilution of the patient DNA. All testing was done in triplicate to determine intra-assay concordance. Lastly, testing was repeated to verify inter-assay repeatability.
Results: 5–7 sets of primers were designed for each patient sample. Real time PCR data were used to select an optimal primer set for each patient. Optimal primer sets, selected on the basis of specificity, were then tested to determine sensitivity. In 6 of 7 patients the optimal primer set achieved a sensitivity of 10−4 or higher. In 5 of the 7 cases test of tonsil never generated a signal that reached threshold even at 50 cycles. In the test for other two patients, tonsil was amplified at Ct > 45. Tests from 5 of the 7 sets did not amplify product from cell lines with known IGH gene rearrangements indicating the specificity of the primer set. 2 sets amplified DNA from 1 cell line each at a very high Ct value.
Conclusion: We designed and developed 7 patient specific MRD tests in times averaging 3–4 weeks. In this study, we were able to design a primer set with a sensitivity of at least 10−4 in 6 of the 7 cases without compromising the specificity. Our experience with this patient set suggests that real time PCR based MRD tests can be developed quickly and efficiently using our methodology to provide assays of both high sensitivity and specificity.
GETPrime: a gene- or transcript-specific primer database for quantitative real-time PCR
