Navigating the misty lands of monocytosis

In this issue of Blood, Baumgartner and colleagues¹ retrospectively apply the recently revised criteria for the definition of chronic myelomonocytic leukemia (CMML) by the World Health Organization (WHO) classification² and the International Consensus Classification (ICC)³ to a large cohort of 5541 patients with monocytosis, including cases with mild monocytosis (ie, absolute monocyte counts between 0.5 and 1.0×10⁹/L), previously referred to as “oligomonocytic” and now classified as CMML. Despite different genetic, immunophenotypic, and transcriptional profiles, no significant differences in survival were found between newly classified and established CMML cases. The authors present the most comprehensive analysis to date of patients with monocytosis in a routine diagnostic setting and provide relevant insight into the implications of the newly proposed classification systems.

For almost the last half century, CMML was defined by the hallmark of peripheral blood (PB) monocytosis ≥1×10⁹/L, first introduced in 1982 as an arbitrary value by the French-American-British classification.⁴ Even with this intrinsic limitation, this cutoff offered an effective framework for capturing patients with myeloproliferation associated with dysplastic features, eventually leading to the recognition of CMML as a myelodysplastic/myeloproliferative neoplasm (MDS/MPN).⁵ Based on recent studies showing that MDS neoplasm with a 0.5 to 1×10⁹/L monocyte count (if ≥10% of the total white blood cells) showed clinical and biological features not dissimilar from CMML,^6,7 both the WHO and ICC systems lowered the cutoff for absolute monocytosis to ≥0.5×10⁹/L (again, despite the potential of bias due to an arbitrary value), thereby incorporating oligomonocytic cases into CMML. In addition, both classification systems recognized proof of clonality as a key criterion for defining CMML. However, the two classification systems now require divergent supporting morphologic criteria. The WHO classification requires the documentation of dysplasia in at least one lineage, whereas the ICC requires bone marrow (BM) hypercellularity due to myeloid proliferation and PB cytopenia as additional criteria. Finally, both systems introduced precursor states relevant to CMML. For the WHO it is included in the broad definition of clonal cytopenia of undetermined significance (CCUS), whereas the ICC has a specific category of clonal monocytosis of undetermined significance (CMUS).^2,3 The figure depicts the key changes introduced by the WHO and ICC in the classification of CMML and premalignant monocytosis.

Baumgartner and colleagues show a high rate of concordance between BM dysplasia (required criterion for the WHO) and PB cytopenia (required by ICC). Conversely, the additional ICC criterion of a CMML-typical BM morphology was found in only a fraction of WHO cases.¹ Then, the authors analyzed a cohort of 2130 well-annotated oligomonocytic cases and found that 17% and 11% were classified as CMML according to WHO and ICC, respectively. It is worth noting that 1.5 times more oligomonocytic cases were classified as CMML based on the WHO as compared with the ICC criteria. This difference was mainly due to the aforementioned BM requirements. A closer inspection of reclassified oligomonocytosis cases reveals that approximately 90% of them had been previously classified as MDS according to the WHO 2017, whereas oligomonocytic cases exclusively classified as CMML by the WHO either remained within MDS or fell into CMUS according to the ICC. Not surprisingly, genetic analysis of the newly classified CMML cases according to the WHO criteria showed strong enrichment in MDS-specific alterations.¹ Taken together, these results suggest that the WHO supporting morphologic criterion of dysplasia resulted in a higher proportion of MDS being reclassified as CMML, but it appears to be less specific in capturing myeloproliferative features at both the clinical and molecular level. Conversely, ICC criteria appear less sensitive but significantly more specific in capturing a proliferative footprint. Additional studies are therefore warranted to clarify whether the MDS cases with mild monocytosis reclassified by the WHO will show clinical trajectories consistent with MDS/MPN or more consistent with bona fide MDS. Despite these differences, Baumgartner and colleagues provide evidence that the reclassified CMML cases showed comparable survival to classical CMML cases after adjusting for known prognostic variables,⁸ corroborating the notion of a continuum between MDS and MDS/MPN and supporting the identification of meaningful disease subgroups based on specific genetic signatures rather than arbitrary thresholds of monocyte count.

Baumgartner and colleagues also show that morphologic criteria are largely redundant when evidence of clonality can be obtained, further suggesting that the proof of clonality is indeed a sine qua non criterion. This association is clearly shown in cases with overt monocytosis, whereas the association appears less robust in oligomonocytic cases, suggesting some caution when dealing with earlier stages of the evolutionary trajectory of the disease. A recent study exploring monocytosis and its association with clonal hematopoiesis (CH) in community-dwelling individuals showed that older individuals with monocytosis more frequently carry CH.⁹ However, in only a fraction of subjects with monocytosis is the clonal process driven by a mutational spectra deviating from age-related CH and consistent with early stages in developing malignant myelomonocytic disease, whereas in the large majority of cases CH does not reflect clonal monocytic proliferation. Collectively, these findings support the rationale for introducing clinical precursor states. The available evidence suggests that, in the context of CCUS, a mild monocytosis is highly predictive of MDS/MPN evolution.¹⁰ Although these observations provide grounds for recognizing a specific monocytic precursor state, identification of either CCUS and CMUS will provide the optimal frameworks for implementing appropriate follow-up before establishing a diagnosis of overt malignancy.

In summary, Baumgartner and colleagues provide a robust basis for implementation of the newly proposed revisions to classification of myeloid neoplasms with monocytosis, and offer relevant clinical and biological insights on newly classified cases. Their findings also support the addition of clinical and biological studies to identify within the spectrum of MDS and MDS/MPN entities based on molecular rather than phenotypic profiles, as well as to segregate premalignant and early malignant precursor states, which is critical for early identification of malignant monocytosis.

Conflict-of-interest disclosure: L.M. has no relevant conflicts of interest related to this work.