Cytopenias such as anemia and thrombocytopenia are common complications in patients with myelodysplastic syndromes (MDS), but their severity and time course are heterogeneous. New therapeutic agents such as luspatercept are available that can have some modest beneficial effects on cytopenias. Early interventions with these newer agents are being evaluated, but additional data are needed to determine whether early intervention improves long-term patient outcomes.

There is a lot of heterogeneity in the trajectory of cytopenias in patients with MDS, so it is challenging to take clinical trial data and apply them to patients in the clinic. I have patients with a hemoglobin level of approximately 8 g/dL who are unaffected by it. They are as active as they have ever been, and I see them every few months to make sure that it is stable. This approach is less of a burden on these patients than having to come in for an injection every 3 weeks. In contrast, I have other patients whose hemoglobin is hovering around 9 or 10 g/dL, but they are terribly fatigued. This can also be challenging because you cannot necessarily pin the fatigue on anemia.

There are probably a lot of considerations other than just oxygen delivery based on hemoglobin levels that impact a person’s lived experience with MDS. It is difficult to know whether intervening will provide clinical benefit, especially in those with borderline blood counts. And one must consider the old dictum, “Who are you treating? Yourself or the patient?”

Because of this, I will often ask my patients to guess what their hemoglobin level is when they come in. Anecdotally, I find that many who can predict it accurately are also the patients who have a “symptom threshold” at a particular hemoglobin level, and who feel better right away after I improve their anemia. But there is another cohort of patients whose hemoglobin estimates are often off from the actual values. For these patients, it can be more difficult to fix any symptoms just by improving the numbers.

For patients with anemia plus moderate thrombocytopenia, I have seen some marginal benefits from using luspatercept. With this agent, for instance, you can get a patient’s platelet count to improve from 20,000/µL to the range of 30,000/µL to 50,000/µL, where you are not as worried about bleeding and you do not have to see them as often. This was reported in a subgroup of patients with pancytopenia from the MEDALIST trial, where treatment with luspatercept was associated with small improvements in thrombocytopenia.

In an era where we are starting to use therapies earlier, we also have to think about the long-term impact of starting that treatment. If you are going to intervene before a patient has a full manifestation of disease, you want to make sure that you improve that patient’s life. So, I think that if you are going to treat someone before they become anemic, we should also think about evaluating de-escalation models of therapy so that the patient has periods away from health care. Otherwise, all you have done is given them 2 more years of coming in every 1 to 3 weeks for treatment. An example is the Sintra-REV study, where patients with del(5q) MDS received time-limited low-dose lenalidomide therapy for 2 years. The study showed that time-limited lenalidomide prolonged time to transfusion dependence vs placebo; however, further follow-up is needed to determine whether treatment improves long-term outcomes.