To the Editor:
Professor Morley was correct in pointing out in our Perspective we focused on analysis of Poisson noise, a type of sampling error [1]. Indeed, it is sometimes useful to distinguish between errors in sampling cells from a sub-population and errors in sampling a sub-population from the entire population of cells. The underlying mechanisms for these two types of sampling errors could be similar. Although in our Perspective we did not dwell on spatial distribution of leukaemia cells, we now caution an uneven spatial distribution could also be due to Poisson noise, as exemplified by R. D. Clarke’s classic spatial analysis of the distribution of flying-bomb attacks in London during WWII [2]. It is also important to recognise any sampling error at the cell-count level cannot be salvaged by lysis of the sampled cells for subsequent nucleic acid analysis such as quantitative real-time polymerase chain reaction (RT-qPCR) or next-generation sequencing (NGS).
Some leukaemia treatment protocols do call for measurable residual disease (MRD)-testing at a time when collecting a large number of bone marrow cells is not always feasible [3, 4]. The ideal scenario is of course having a multi-parameter flow cytometry (MPFC)-based MRD-test declared positive only if ≥5 × 10E+5 cells are analysed and if ≥50 cells are positive, but oftentimes physicians need to make decisions under non-ideal conditions. Should consideration of sampling errors affect the treatment plan for a person? We believe it should, but not until more validation studies are conducted. We agree a global platform such as EuroMRD would be the right venue for advancing proper usage of MRD-tests.
Professor Morley argued that decisions based on the conventional MRD values will optimise treatment for the group as a whole. We disagree. The conventional MRD value is not the mean or median estimate of true MRD. Rather, when conventional MRD is zero, it is the optimist’s rosy estimate of true MRD assuming all such patients have near-zero leukaemia cell. Making decisions based on such false optimism would not optimise treatment for the group as a whole, not to mention some of the individual patients. Perhaps even a hospital administrator should consider including \({{{{{{\rm{MRD}}}}}}}_{{{{{{\rm{worst}}}}}}\_{{{{{\rm{case}}}}}}}\) as one of her benchmarks for evaluating treatment efficacy as a whole.
Finally, in our cohort of children with acute lymphoblastic leukaemia treated on CCCG-ALL-2015, \({{{{{{\rm{MRD}}}}}}}_{{{{{{\rm{worst}}}}}}\_{{{{{\rm{case}}}}}}}\) identified sub-groups of children with poorer relapse-free survival even though their conventional MRD-test results were nearly all zero (Fig. 1). Nonetheless, like others we avoid relying on composite end points that could have heterogeneous make-up of adverse events [5]. In our opinion, it is crucial for an MRD researcher to clarify the impact of an MRD-testing result on the cumulative incidence of relapse itself.
References
Feng Y, Qi S, Liu X, Zhang L, Hu Y, Shen Q, et al. Have we been qualifying measurable residual disease correctly? Leukemia 2023;37:2168–72.
Clarke RD. An application of the Poisson distribution. J Inst Actuaries 1946;72:481.
Jeha S, Pei D, Choi J, Cheng C, Sandlund JT, Coustan-Smith E, et al. Improved CNS control of childhood acute lymphoblastic leukemia without cranial irradiation: St Jude total therapy study 16. J Clin Oncol 2019;37:3377–91.
Yang W, Cai J, Shen S, Gao J, Yu J, Hu S, et al. Pulse therapy with vincristine and dexamethasone for childhood acute lymphoblastic leukaemia (CCCG-ALL-2015): an open-label, multicentre, randomised, phase 3, non-inferiority trial. Lancet Oncol 2021;22:1322–32.
Walia A, Tuia J, Prasad V. Progression-free survival, disease-free survival and other composite end points in oncology: improved reporting is needed. Nat Rev Clin Oncol 2023;20:885–95.
Acknowledgements
I acknowledge support from the Institute of Hematology, Chinese Academy of Medical Sciences (IHCAMS). I thank Yahui Feng, Wei Zhang and Saibing Qi for assistance in preparing this typescript.
Funding
Supported, in part, by grants from the Chinese Academy of Medical Sciences (CAMS) Innovation Fund for Medical Sciences (2021-I2M-1–001 and 2022-I2M-2–003).
Author information
Authors and Affiliations
Contributions
JC prepared the typescript and takes responsibility for the content.
Corresponding author
Ethics declarations
Competing interests
The author declares no competing interests.
Additional information
Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.
About this article
Cite this article
Chen, J. Response to comment on Have we been qualifying measurable residual disease correctly?. Leukemia 38, 219–220 (2024). https://doi.org/10.1038/s41375-023-02088-4
Received:
Revised:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1038/s41375-023-02088-4