Close-up on abnormal monocyte morphology in peripheral blood smears

Karen Lusky

March 2021—A recently published study of 90 patients found significant numeric and atypical white blood cell morphologic changes associated with SARS-CoV-2 infection that differed between mild and severe disease. The changes also differed for critically ill patients with and without the infection (Pozdnyakova O, et al. Am J Clin Pathol. 2021;155[3]:​364–375).

Olga Pozdnyakova, MD, PhD, medical director of the hematology laboratory at Brigham and Women’s Hospital, reported the findings of the study in her CAP20 virtual presentation on monocytosis, in a session on a morphology-based approach to hematopoietic neoplasms presenting with an abnormal WBC differential.

Dr. Pozdnyakova, who is also associate professor of pathology at Harvard Medical School, spoke first on neutrophilia (see CAP TODAY, February 2021) and then on monocytosis. In the latter presentation, she shared, in addition to the results of the latest study, a non-SARS-CoV-2 case of a man who presented with persistent monocytosis.

When the pandemic began, she said in a recent interview, technologists were leaving a lot of peripheral blood smears for pathologists to review because they began to observe striking changes in the monocytes of COVID-19 patients. “We didn’t know they were patients with COVID, but when I traced it back, the unifying thinking was that all had been diagnosed with COVID-19. This is what prompted the study—striking changes with large coalescing vacuoles in monocytes.”

Dr. Pozdnyakova and colleagues characterized 90 patients diagnosed with COVID-19 (non-ICU and ICU). Their peripheral blood revealed many reactive changes in the form of monocyte and lymphocyte vacuolization, atypical lymphocytes, large granule lymphocytes, myeloid left shift, toxic granulation of neutrophils, and vacuolization of neutrophils.

“More mild disease was associated with more prominent monocyte changes and the presence of atypical lymphocytes,” she reported, “while more severe disease was associated with more prominent myeloid left shift. Patients who succumbed to the disease and demonstrated a more severe clinical course lost prominent monocyte vacuolization and became more neutrophilic and more lymphopenic with time.”

“Despite the difference in the monocyte morphology,” she continued, “there was no difference in monocyte count between the less severely affected and more severely affected patients.”

Why the monocytes tend to lose this striking vacuolization is unknown, Dr. Pozdnyakova tells CAP TODAY. “We don’t even know what these vacuoles are, but one of the theories is that it’s possible there are different monocyte subsets because monocytes are not all the same. They are all immunologically different. And maybe there are different monocyte subsets in patients who do better with COVID-19 or do worse with COVID-19. And the ones who do better present with the striking morphologic features.”

She and her colleagues acknowledge that an analysis of morphologic changes in WBC “is operator dependent and as a result could be subjective.

“In an attempt to standardize morphologic review of peripheral blood smears” in SARS-CoV-2 patients, they write, “we have performed an exploratory analysis of morphology-associated research parameters, which are measured but not reported with every CBC with automated differential at our institution. Our analysis revealed significant differences in the research parameters related to morphology of neutrophils, monocytes, and lymphocytes between COVID-19-positive and COVID-19-negative ICU patients.”

Their experience with these research parameters is scant, they say, thus limiting the interpretation of the findings. “However,” they write, “the presence of different parameter values between COVID-19-positive and COVID-19-negative patients and the absence of such difference between COVID-19-positive ICU and non-ICU patients suggests that some of these changes could be attributed to the SARS-CoV-2 virus.”

“These are the parameters that are not reportable currently, and we don’t quite know what some of them mean,” Dr. Pozdnyakova tells CAP TODAY, “but I think some of them measure morphologic changes in different patients with different diagnoses.” If the parameters were proved to be associated with morphologic changes, “there would be a more objective way to measure changes in cell morphology and correlate them with those objectively measured parameters by hematology analyzers,” she says. “It has the potential to help us better screen patients who will need further hematological evaluation.”

Dr. Pozdnyakova and coauthors conclude: “Hospitalized patients with COVID-19 should undergo a comprehensive daily CBC with manual WBC differential to monitor for numerical and morphologic changes predictive of poor outcomes and signs of disease progression.”

[dropcap]M[/dropcap]onocytosis is defined as more than 0.8 × 10⁹/L or, in cases of chronic myelomonocytic leukemia, more than 1.0 × 10⁹/L. Like neutrophils, they can demonstrate reactive changes. “Instead of referring to dysplastic monocytes, we usually talk about abnormal monocytes,” Dr. Pozdnyakova said in her presentation. In cases that present with monocytosis, immature monocytes include not only monoblasts but also promonocytes, which are counted as blast equivalents and included in the final blast count.

Monocytosis can be primary (neoplastic), which is almost always the case in people who present with persistent monocytosis, or secondary (reactive). Secondary etiologies are more common than primary and can be associated with inflammation, infection, autoimmune disorders, and drugs such as corticosteroids and growth factors. Monocytosis can be associated with a hematologic or non-hematologic malignancy.

Among neoplastic causes of monocytosis, chronic myelomonocytic leukemia (CMML) is the most common, Dr. Pozdnyakova said. Other causes: juvenile myelomonocytic leukemia, acute myeloid leukemia with monocytic differentiation, CML with p190 fusion, and myeloid neoplasm with rearrangements of PDGFRA, PDGFRB (this is the most common one, she said), FGFR1, or PCM1-JAK2. “And monocytosis could also be a sign of progression of Philadelphia-negative myeloproliferative neoplasm.”

Monocytes are the largest white blood cells, with abundant cytoplasm, which could be agranular or contain evenly distributed, fine azurophilic granules and small vacuoles. The nuclear chromatin is condensed. “It’s a little less condensed than we typically see in the neutrophils,” Dr. Pozdnyakova said, “but it’s still condensed, and we should not see any nucleoli.”

In her CAP20 talk, Dr. Pozdnyakova described a 65-year-old man who presented with persistent monocytosis. The CBC that came to their attention showed neutrophilia with an absolute neutrophil count of greater than 13,000, monocytosis approaching 7,000, significant anemia, and a low platelet count.

The patient was a chronically ill person who was fatigued, complained of night sweats, and had abdominal distension caused by splenomegaly. Examination revealed that the patient had an absolute monocyte count of over 5,000 per microliter for at least four months. “So at this point our suspicion for malignancy is very high,” she said.

In Fig. 1 is the patient’s peripheral blood smear showing monocytosis with highly atypical monocytes. “There is a higher than normal nucleus-to-cytoplasmic ratio with abnormal nuclear shapes and chromatin patterns,” Dr. Pozdnyakova noted. She pointed to twisted nuclei (upper left corner) and to some nuclei that are not connected (left side of the box in lower left corner).

The chromatin patterns of some of the nuclei are very much condensed, she said. “But some are a little less mature and you can even see some small nucleoli” (Fig. 1, top right). “Cytoplasmic granules are not very prominent,” and only a little bit of cytoplasmic vacuolization can be seen.

Also seen is dysplasia in granulocytes, characterized by hypersegmentation, as well as hypergranularity and the presence of small vacuoles, Dr. Pozdnyakova said (Fig. 1, two granulocytes on far right).

CMML is the most common of the neoplastic disorders associated with monocytosis, but several other diagnoses need to be excluded using clinical history, genetic findings, and morphologic review. “The patient’s age excludes a juvenile myelomonocytic leukemia, and because we do not have a history of prior Philadelphia-negative myeloproliferative neoplasm, we can exclude progression”—essential thrombocythemia, primary myelofibrosis, and polycythemia vera.

“So we are left with CMML, AML with monocytic differentiation, chronic myeloid leukemia with p190, or myeloid neoplasms with eosinophilia, especially the ones associated with PDGFRB rearrangement,” Dr. Pozdnyakova said. “Genetic testing is always indicated in such cases,” and the patient’s karyotype (46, XY[20]) was normal, helping to exclude CML, characterized by translocation 9;22.

The testing also helped to exclude PDGFRB-rearranged neoplasm where abnormalities are usually seen in 5q31-32, most commonly t(5;12). “And the PDGFRB-rearranged neoplasm usually, but not invariably, presents with eosinophilia, which we didn’t see in this case,” Dr. Pozdnyakova said.

The next-generation sequencing results for the patient identified four somatic mutations: TET2, ASXL1, CBL, and KRAS. Frequencies are 49.7 percent for TET2, 43.9 percent for ASXL1, 19.6 percent for CBL, and 15.2 percent for KRAS, “suggesting that the two latter mutations represent a subclonal event.”

Although the mutations confirm the diagnosis of a myeloid neoplasm and, in conjunction with the clinical presentation of persistent monocytosis and dysplasia, is compatible with CMML, “those mutations cannot help us distinguish between CMML stages, or even AML with monocytic differential, since the distinction between these entities relies on our ability to recognize immature monocytic cells solely based on morphology,” Dr. Pozdnyakova said.

The classification of CMML stages is based on the percentage of blasts in the bone marrow and peripheral blood. CMML-0 is defined as less than two percent blasts in peripheral blood or less than five percent blasts in bone marrow. CMML-1: two to four percent blasts in peripheral blood or five to nine percent blasts in bone marrow. CMML-2: five to 19 percent blasts in peripheral blood or 10 to 19 percent blasts in bone marrow. “And acute myeloid leukemia with monocytic differentiation requires equal to or greater than 20 percent blasts” in bone marrow or peripheral blood.

In cases presenting with monocytosis, blasts include not only monoblasts but also promonocytes, referred to as blast equivalents. “There are no immunophenotypic markers that can help us with the distinction between mature and immature monocytic forms—for example like we see with CD34 or CD117 in the myeloblast. And correct CMML classification that changes the prognosis, which is worse with the high blast count, only relies on our ability to recognize immature forms,” she said.

“In addition to the blast count, other factors that change prognosis are high white blood cell count and significant anemia or thrombocytopenia at presentation, presence of abnormal karyotype, and certain mutations, such as ASXL1, that we saw in our patient.”

In clinical practice, it’s not necessary to differentiate between monoblasts and promonocytes (Fig. 2) because both are counted as blasts. “Both immature cell types show the same cytoplasmic morphology with deeply basophilic cytoplasm that often contains fine granules or may contain cytoplasmic vacuoles,” she said.

Dr. Pozdnyakova said she uses nuclei shape to distinguish between monoblasts and promonocytes. Monoblasts have round or slightly indented nuclei, she said, while promonocytes have more indented nuclei and sometimes irregularly shaped nuclei. “However, independent of the nuclei shape, they all have immature chromatin quality and very often they contain nucleoli.”

Fig. 3 is a peripheral blood smear from a patient with CMML-1 with four percent circulating blasts. “You can still see plenty of mature myeloid and monocytic cells in the background,” Dr. Pozdnyakova said, noting also the blast (arrow).

In contrast, Fig. 4 is a peripheral blood smear from a case of acute myeloid leukemia with monocytic differentiation where all cells are immature. There is “one possible myeloblast,” she said (top of image), and the rest of the cells are promonocytes.

Microgranular variant of acute promyelocytic leukemia (Fig. 5) is a look-alike for acute myeloid leukemia with monocytic differentiation, Dr. Pozdnyakova warns, “since the promyelocytes in this AML subtype rarely contain Auer rods, and the cells are large with moderate to sometimes abundant agranular cytoplasm and irregularly shaped immature nuclei.”

In the cases of microgranular APL, she said, the key is to recognize cells with bilobed or butterfly nuclei, as seen in Fig. 5, that are characteristic for the microgranular APL variant. “It is essential to recognize acute promyelocytic leukemia since patients should start with ATRA treatment to avoid development of DIC. The correct diagnosis in such cases relies on recognition of atypical promyelocyte morphology and confirmatory ancillary testing, which includes cytochemical and immunophenotypic analysis, and cytogenetic analysis that shows the presence of 15;17 with rearrangement between PML and RARA genes.”

Cytochemical staining is a rapid and helpful way to differentiate between those cell types, she said (Fig. 6). Atypical promyelocytes are strongly positive for myeloperoxidase even in the absence of Auer rods, she said. “The staining often obscures the nucleus, while immature monocytic cells are negative for MPO but positive for nonspecific esterase, and it can be variable intensity of nonspecific esterase.”

Flow cytometry immunophenotyping (Fig. 7) can help assign the cell lineage. The top row in Fig. 7 is a case of acute promyelocytic leukemia; the bottom row is acute myeloid leukemia with monocytic differentiation. “Atypical promyelocytes are located in the granulocyte gate with lower CD45 and high side scatter, while atypical monocytes or immature monocytes are located in the monocyte gate with a little high CD45 and lower side scatter,” Dr. Pozdnyakova said. “Promyelocytes are positive for the myeloid marker CD13 and CD33, and strongly positive for myeloperoxidase stain, similar to the cytochemical stain, and then negative for CD34 and HLA-DR, while immature monocytes are positive for HLA-DR, along with monocytic markers CD11b and CD64.” CD34 is almost always negative on an immature monocyte population.

In this case, the myeloblast population was also present (seen in blue). “Those cells are positive for HLA-DR and CD34. They’re located in the traditional blast gate, and they’re positive for myeloid marker CD13 and dim CD33. However, they’re negative for the monocytic markers.”

Dr. Pozdnyakova reported that their 65-year-old patient was diagnosed with CMML-0 based on his having no immature cells in peripheral blood and two percent immature cells in the bone marrow. However, because the patient was symptomatic and had a high WBC count, anemia, low platelet count, and ASXL1 and KRAS mutations, “he was assigned to a high-risk category and started on the cytoreductive therapy with hydroxyurea.”

[dropcap]I[/dropcap]n summing up, Dr. Pozdnyakova said monocytic differentiation is defined as large cells with abundant cytoplasm, with or without fine granules and with an absence of Auer rods. By cytochemistry, those cells are positive for nonspecific esterase with variable intensity. Immunophenotype (shared between mature and immature monocytes) is the presence of the monocytic markers (CD11b, CD14, CD64, and CD68 are most helpful). “And there could be some variances, such as variable expression of CD34, CD117, or CD56.”

In Fig. 8 are the genetically defined acute myeloid leukemia types often associated with monocytic differentiation.

Persistent monocytosis (>1 × 10⁹/L, more than three months) is almost always neoplastic, especially when accompanied by thrombocytopenia or anemia or both. “You need to always exclude CML, especially CML with p190, and PDGFRA, PDGFRB, FGFR1, and PCM1-JAK2-associated neoplasms, which usually, but not always, present with eosinophilia.” PDGFRB can present as CMML, and in those cases karyotype and FISH are essential.

“You need to recognize atypical monocytes and assess dysplasia in granulocytes to arrive at the correct diagnosis. And remember that the immature form of blast equivalents includes monoblasts and promonocytes,” and recognizing them requires knowing their morphology well. Finally, next-generation sequencing is a must, she said, when neoplastic monocytosis is suspected to confirm clonality and assess prognosis. 

Karen Lusky is a writer in Brentwood, Tenn.