Next Generation Sequencing-Based Methods Yield Reduced Inter-Laboratory Variability Compared to Short Tandem Repeat-Based Methods in ASHI Engraftment Monitoring Proficiency Testing Results

Aim: The ASHI Engraftment Monitoring (EMO) Survey is designed to evaluate a laboratory's ability to determine the proportion of DNA or cells from two individuals in a mixture of their blood samples as a model of testing for hematopoietic chimerism. We aimed to compare the inter-laboratory variability between the two most common submitted methods, NGS and STR, by analyzing ASHI EMO survey results.

Methods: EMO results were extracted from the ASHI Proficiency Testing Data Center from 2018-2024. To evaluate potential variability differences at different ranges of chimerism, results were stratified into three categories based on the average minor fraction: 5 ≤ 10%, 11 ≤ 20%, and 21 ≤ 50%. Individual results were excluded if the method used was not defined or if a result was a clear sample swap or donor/recipient value inversion. Data from 2018-2024 were analyzed to assess trends in reported EMO methods and analysis of inter-laboratory variability was performed using specimens from 2021-2024 to gather a sufficient replicate size for each method per sample. A weighted average standard deviation (SD) was calculated to account for variations in the numbers of results submitted by method.

Results: The number of participants in the EMO survey has increased from 49 laboratories in 2018 to 66 in 2024. While STR has been the most common method, the number of STR results has recently declined from 46 to 39 participants while the number of participants utilizing NGS-based methods has increased from 2 to 25. NGS-based results had the lowest average SD across all ranges of minor fractions, 0.75, 0.96, and 1.47 for 5 ≤ 10%, 11 ≤ 20%, and 21 ≤ 50%, respectively, while STR had SD values of 1.44, 1.63, and 2.95 for the same respective categories (Table 1; 5 ≤ 10% fraction represented in Figure 1).

Conclusion: Compared to STR testing, NGS-based results displayed less inter-laboratory variability across all categories of minor fractions evaluated. This difference may be clinically beneficial as NGS-based methods could be more suitable to detecting incremental changes in chimerism to more accurately detect early relapse and more closely evaluate responses to treatment.