Submitting Author: Gerbi, Rada, MD
Institution: Henry Ford Health System
Additional authors:Kristin Hunt Karner, MD
Session: Therapy-related myeloid neoplasms

This bone marrow biopsy is from a 72-year-old male with rheumatoid arthritis treated with methotrexate for about six months. He developed pancytopenia and a bone marrow biopsy was performed (see attached images).

Due to sustained cytopenias, methotrexate was discontinued and peripheral blood counts improved. The patient received no other chemotherapy. Another bone marrow biopsy was performed two years after cessation of methotrexate, revealing complete resolution of the previous dysplastic features and blasts.

The patient has also had serum protein electrophoresis showing IgG lambda monoclonal protein of 0.1-0.2 g/dl, not changing significantly over time.

Random bone marrow biopsy and aspirate, left iliac crest. Fixed in Bouins and decalcified. Core length 0.9 cm. Adequate particles on aspirate smears.

The marrow is hypocellular with a myeloid: erythroid ratio of approximately 1:1. The myeloid lineage is left-shifted with 13% blasts. The erythroid lineage is also left-shifted with moderate dyserythropoiesis and about 10% ring sideroblasts. The majority of megakaryocytes have unremarkable morphology.

Images/slides are also included of the follow-up biopsy two years later, showing normalization of the cellularity, blast count and dysplastic features. Incidentally, there was evidence of a plasma cell dyscrasia consistent with monoclonal gammopathy of undetermined significance (MGUS) that was also present in the original biopsy (3-4% monoclonal plasma cells).

Flow cytometry confirms increased myeloblasts, left-shifted granulocyte maturation profile and decrease in side scatter of granulocytes.

CD34 and CD117 immunohistochemical stains performed on the core biopsy also show the increased blast population, consistent with aspirate smear counts.

MDS FISH panel was negative for any cytogenetic abnormalities.

(Two years later, MDS FISH panel is still negative and conventional chromosomal analysis showed 46,XY in 20 out of 20 cells.)

Not performed.

It is well known that both rheumatoid arthritis and the therapies used to treat this disease can cause morphologic findings in the bone marrow that can mimic MDS. Usually these changes are fairly mild. This case is unusual in that it shows a large blast population as well as significant dysplastic changes including ring sideroblasts, and was initially thought to represent a myelodysplastic syndrome. The lack of complete maturation in the myeloid lineage in the context of a hypocellular marrow may have been a clue that these blasts represented a normal regenerating myeloid lineage after cytotoxic insult. Even though the findings present in the initial biopsy would technically meet WHO criteria for RAEB-2, MDS is ultimately a diagnosis of exclusion. This case highlights a potential diagnostic pitfall in patients with autoimmune diseases such as rheumatoid arthritis, and for patients receiving chemotherapeutic agents such as methotrexate. Obtaining a thorough clinical history is essential before a diagnosis of MDS can be made in the context of normal cytogenetic results.

Hypocellular marrow with increased blasts and erythroid dysplasia including ring sideroblasts, most consistent with effects of methotrexate therapy for rheumatoid arthritis.

Hypocellular marrow with increased blasts and erythroid dysplasia, consistent with effects of methotrexate therapy for rheumatoid arthritis

Case 57