Impact of Comorbidities on Prognosis of Elderly Patients with Acute Myeloid Leukemia Who Receive Hypomethylating Agents

BACKGROUND: Many efforts have been made in the attempt to address the conundrum question of fitness definition ad prognosis prediction in elderly acute myeloid leukemia (AML) patients. Parametric definitions are expected to give an advantage in patient stratification; however, clinical examination remain de facto pivotal to formulate therapy decisions and frequently the comorbidities are empirically evaluated.

METHODS: We conducted a multicenter study collecting baseline comorbidity, laboratory data, CTCAE 4.0.3 adverse events (AE), and outcome of elderly patients (>65 years old) with new onset AML who received hypomethylating agents as 1st line therapy. We tested the impact on prognosis of baseline clinical and biological risk factors. Furthermore, we evaluated a score - acute myeloid leukemia-composite model, AML-CM (Mukherjee et al, 2017) - developed in chemotherapy-eligible patients, that accounts for baseline comorbidities, laboratory parameters, age and cytogenetic-molecular risk. The study was approved by local Ethical Authority (316/2019/Oss/AOUBo).

RESULTS: We collected data from 131 consecutive elderly patients who received 1 ^st line HMAs between January 2008 and January 2021. Patients had a median age of 76 years (IQR 72 -79). Seventy-seven out of 131 patients (58.8%) had de novo AML, 32/131 (32.8%) had secondary AML, and 11/131 (8.4%) had therapy-related AML. Out of 123 evaluable patients, 43 (34.9%) had complex karyotype, 1 (0.8%) inv(16), 59 (48.4) normal karyotype, 18 (14.7%) other alterations; 8/108 patients harbored FLT3 ITD mutation (7.4%, 23 not tested), 12/101 NPM1 mutation (11.9%, 30 not tested). Based on these data, 111 patients were evaluable for ELN2010 risk stratification; 9 over 111 patients (8.1%) were stratified in the low risk, 42/111 (37.8%) in intermediate-1 risk, 17/111(15.3%) in intermediate-2 risk, and 43/111 (38.7%) in high-risk class.

As expected, most of the patients had at least one comorbidity. Particularly, baseline arrhythmia was present in 29/130 (22.1%, 1 no data), cardiovascular comorbidity in 20/130 (15.4%, 1 no data), diabetes in 20/131 (15.3%), cerebrovascular comorbidity in 11/131 (8.4%), kidney disease in 15/130 (11.5%, 1 no data), lung chronic disease 19/130 (14.6% 1 no data), hypoalbuminemia in 25/111 patients (22.5%, 20 no data).

With a median follow up of 28.2 months, median overall survival (OS) of the entire cohort was 15.8 months (95% C.I. 11.2-19.4). We confirmed that patient who obtained a response (complete remission, partial response, hematological improvement) after 2 months of therapy had the best OS (figure A, median OS of 21 months for responders vs 7.4 months for non-responders, p <.001). Interestingly, lung chronic disease (median OS 6.6 months in affected vs 16.5 months in non-affected, p=.013) and hypoalbuminemia (median OS 7.4 months in affected vs 18 months in non-affected p<.001) confer significantly diminished OS. ELN2010 score impacted prognosis (median OS of 8.4 months for favorable, 23.4 months for int-1, 11.1 for 1int-2 and 6.5 months for high-risk, p=.004).

To test the impact of comorbidities combined with cytogenetic and molecular risk, AML-CM was used. Our results indicate that AML-CM score was able to stratify prognosis in elderly patients receiving frontline HMAs (figure B, median OS in score group 1: 29.7 months, score group 2: 16.5 months, score group 3: 11.2 months, score group 4: 6.6 months, p=.038). The worse prognosis of patients with higher AML-CM score, which includes patients with increased baseline comorbidities, may be explained with a higher incidence of AEs (84.55, 116.01, 131.45, 229.3 events for 100 patients per year for score group 1,2,3, and 4, respectively) and infections (53.80, 55.10, 85.95, 140.13 events for 100 patients per year for score group 1,2,3, and 4, respectively), in patients with higher baseline comorbidities.

CONCLUSION: In this study we found that baseline comorbidities, captured by AML-CM score, may define prognosis of elderly patients receiving 1st line HMAs; parametric comorbidity scores may improve our ability to predict outcome and tailor interventions. The impact of comorbidity on OS may be increased with novel and more aggressive therapy. For this reason, specific studies on functional fitness tests and geriatric assessments are highly warranted in patients receiving HMAs plus venetoclax.

This work was supported by Bologna AIL. CP and AC shared last authorship.

Figure 1

View largeDownload slide

Figure 1

View largeDownload slide

Close modal