Dual antibody immunohistochemistry: an efficient and sensitive tool for the detection of residual disease in chronic lymphocytic leukemia

Highly effective treatments for chronic lymphocytic leukemia (CLL) have the potential to reduce significant tumor burden to single cells and therefore require sensitive tools to assess for minimal residual disease (MRD) in bone marrow (BM) biopsies. Flow cytometry (FC) is the current gold standard for detection of MRD, but requires a specialized facility, specific antibody panels, collecting hundreds of thousands-to-millions of cells, and personnel with expertise in the analysis and interpretation of FC MRD data, which may not be feasible in many small laboratories. Dual-antibody immunohistochemistry (DA-IHC) can identify abnormal populations better than morphology alone, but its correlation with FC assessment is not known. Our aims are to characterize the efficacy of DA-IHC in assessing BM samples post-treatment and to compare results with FC to evaluate for residual disease in CLL. We collected 2-year post-therapy data from 33 CLL patients on two treatment protocols, chemoimmunotherapy (CIT) and single agent ibrutinib (IB), as well as BMs from 10 healthy volunteers as morphologic controls. BM biopsy specimens were examined for the presence or absence of CLL based on morphologic evidence of lymphoid infiltration and aberrant co-expression of CD5 and PAX5 DA-IHC. FC using a standard CLL antibody panel was performed in parallel. All IB patients had residual disease detected by DA-IHC and FC, although five patients (22%) were morphologically negative by routine hematoxylin and eosin (H&E) stain. Those without overt morphologic evidence of disease showed DA-IHC-positive interstitial single cells and were positive for residual disease by FC. The assessments by DA-IHC and FC were significantly correlated ( p = 0.004). Four patients (40%) treated with CIT were morphologically negative by H&E, and two of these had no detectable CLL by either DA-IHC or FC. The other two patients had low-level disease detected by both DA-IHC and FC. The MRD levels identified by DA-IHC were correlated with those by FC ( p < 0.0001). We characterized the efficacy of DA-IHC in assessing CLL post-treatment and found assessments by DA-IHC and FC were statistically significant in both the IB- and CIT-treated patients. These findings suggest DA-IHC could accurately detect residual disease in cases lacking morphologic evidence by H&E alone. DA-IHC may be a useful tool in current practice as another sensitive and efficient method to assess for MRD in CLL.