Pediatric Autoimmune Hemolytic Anemia Is Associated with a High Incidence of an Underlying Immune Disorder and High Mortality Rate

Background: Autoimmune hemolytic anemia (AIHA) is a rare pediatric autoimmune disease in which there is limited guidance for evaluation and treatment and a poor understanding of the characteristics that influence prognosis and clinical course.

Aim: To characterize the clinical features, laboratory findings, and treatment outcomes of children with AIHA over a 10-year period.

Methods: An IRB-approved multicenter observational cohort study included patients (pts) from 15 institutions from the ITP Consortium of North America (ICON) who were 3 mo to 21 y and had hematology consultation between Jan 2011 and Dec 2020 with a diagnosis of AIHA. Post-stem cell transplant AIHA was excluded. Demographic, medical history, laboratory, and treatment data were collected. Continuous variables are summarized as mean (SD) or median (range) and compared with a two-sided t-test. Categorical variables are summarized as counts or percentages and compared with a two-sided chi-squared test.

Results: A total of 402 pts were included: median age at diagnosis 7 y (n=397, range: 0-22) with median length of follow up 2.7 years (n=390, range: 0.1-19.2). 45% (180/398) were female. Warm AIHA (wAIHA) was the predominant subtype in 64% (252/395). Most pts (58%, 217/374) had a single AIHA episode that remitted; others had two episodes that remitted (8%, 29/374), multiple relapsing episodes (7%, 26/374), chronic course that remitted (13%, 48/374), or chronic ongoing course (14%, 54/374). Cold agglutinin disease (CA) was present in 12% (47/395), paroxysmal cold hemoglobinuria (PCH) in 6% (24/395), and mixed/unspecified AIHA in 18% (72/395).

Secondary AIHA was common with 15% (57/389) with an underlying primary immunodeficiency, most commonly CVID (49%, 28/57), ALPS (33%, 19/57), and 22q11.2DS (16%, 9/57); and 16% (61/391) with another autoimmune disorder, most commonly SLE (49%, 30/61), neurologic disease (36%, 22/61), or APLA (16%, 10/61). Immune testing was sent in 100% (402/402) and abnormal in 60% (240/402). An immunologist was consulted in 136/342 (40%). Infectious testing was sent in 71% (260/366).

Family history (1^st or 2^nd degree) was positive for immune cytopenias, immunodeficiency, or autoimmunity in 19% (78/402). Genetic testing (sent in 30%, 106/349), was most often a gene panel (67%, 71/106) or WES (28%, 30/106) with pathogenic findings identified in 30% (30/100).

Evans syndrome (ES) was common (37%, 142/385) including ITP (92%, 131/142) and/or immune neutropenia (43%, 57/133) and most likely in those with wAIHA (47%, 113/241) compared with PCH (0%, 0/23) or CA (4%, 2/45). Compared with pts without ES (n=243), those with ES (n=142) were more likely to be older at first episode (9 y vs 5 y, p=0.010), have genetic testing (54% vs 18%, p<0.001), pathogenic gene findings (13% vs 5%, p=0.002), and treatment with steroid-sparing agents at first episode (35% vs 19%, p<0.001). There was no difference in response to steroids at first episode (77% vs 75%, p=0.748) or incidence of another autoimmune disorder (18% vs 14%, p=0.306).

Similarly, those with secondary AIHA were more likely to be older at first episode (8 y vs 6 y, p=0.007) and have wAIHA (77% vs 62%, p=0.035). Secondary AIHA did not impact steroid response (77% vs 74%, p=0.653) or treatment with steroid-sparing agents at first episode (26% vs 24%, p=0.756). Pts with multiple episodes of AIHA were also more likely to have wAIHA (82% vs 52%, p<0.001), genetic testing (50% vs 19%, p<0.001), and treatment with steroid-sparing agents at first episode (34% vs 18%, p<0.001) but did not have a different age at presentation of 1^st episode [8 y vs 5 y, p=0.203] or proportion with pathogenic gene findings (10% vs 6%, p=0.268).

Of those with ongoing follow up, 68/249 (27%) of pts with wAIHA had active disease on treatment, 7/249 (3%) had active disease and were being observed, and 174/249 (70%) were in remission. Overall, mortality was 9.4%. Mortality was not associated with a diagnosis of ES, immunodeficiency or autoimmunity, number of AIHA episodes, or type of AIHA.

Conclusion: AIHA is associated with significant morbidity in children with a high rate of secondary AIHA (~30%), 37% with ES, and 34% with a chronic or relapsing course. In this recently diagnosed cohort, there was a 9.4% mortality rate. These high morbidity and mortality rates require expanded testing, monitoring, and additional treatment approaches in this pediatric population.