Autoimmune hemolytic anemia (AIHA) is a rare disorder characterized by an autoimmune-mediated destruction of red blood cells (RBCs) [1]. Its epidemiology is not well described, but a study in Denmark estimated 1.77 new cases per 100.000 person-years from 2008 to 2016 [2]. Despite being a rare disease, AIHA has a relevant impact on both patients and society and is associated with important healthcare resource utilization, namely due to highly expensive drugs, hospital admissions, and frequent outpatient visits and blood transfusions [3].

The clinical picture in AIHA varies widely, ranging from no symptoms to severe life-threatening hemolysis [4]. AIHA is classified according to the temperature reactivity of the pathogenic antibodies as warm AIHA (wAIHA), cold AIHA, or mixed AIHA [5]. wAIHA, in which autoantibodies target RBC surface antigens at body temperature and trigger hemolysis [6], accounts for the majority of all AIHA cases in the adult population (60–70 %) [1,7]. wAIHA is further categorized as primary (idiopathic) or secondary depending on whether there is an underlying cause. Approximately half of all wAIHA cases are attributed to lymphoproliferative neoplasms, autoimmune and infectious diseases, immunodeficiencies, solid tumors, transplants, or drugs [1,7]. Thrombotic events and infections secondary to immunosuppressive interventions are known complications of wAIHA and can potentially be fatal [8–10].

The direct antiglobulin test (DAT) is the keystone of AIHA diagnosis with its findings being critical to distinguish between the various subtypes [11–13]. In wAIHA, the DAT is typically positive for immunoglobulin (Ig)G only, but it may also show positivity for both IgG and complement component 3d (C3d) or, less commonly, C3d alone. Cold agglutinin disease is characterized by DAT positivity for C3d only. Mixed forms of AIHA demonstrate DAT positivity for both IgG and C3d, with coexistence of warm autoantibodies and a high titer of cold agglutinins. Patients may also present with DAT-negative AIHA, in which testing fails to detect IgG on RBCs despite ongoing hemolysis. This is most commonly due to levels of RBC-bound IgG falling below the detection threshold of standard DAT methods. Less frequently, it may result from low-affinity IgG or from RBC-bound immunoglobulins of other classes, such as IgA or, more rarely, IgM [13].

DAT-negative AIHA occurs in 5–10 % of patients with unequivocal evidence of the disease [7,13]. Consequently, a negative DAT result is insufficient to exclude a diagnosis of wAIHA [7]. The diagnosis is supported by a combination of clinical symptoms, including fatigue, syncope, chest pain, weakness, and exertional dyspnea, and laboratory evidence of hemolysis, specifically anemia, jaundice, and splenomegaly. Characteristic biochemical findings include elevated indirect bilirubin, increased lactate dehydrogenase (LDH), reticulocytosis, and depleted serum haptoglobin levels [5,14–16].

The goal of wAIHA treatment is to reduce the degree of hemolysis, consequentially increasing hemoglobin (Hb) levels and improving symptoms. Management of the underlying cause is crucial in secondary wAIHA, either by treating the associated lymphoproliferative or autoimmune disease or by stopping the drugs that might be causing the disease [17]. Evidence on the treatments used for wAIHA is limited, with a scarcity of randomized prospective clinical trials. As a result, treatment has historically been based on expert opinions and retrospective studies [18–22].

Corticosteroids are the mainstay first-line treatment for wAIHA, with a response rate of 70–80 % [23,24]. However, between 20 and 35 % of patients fail to respond to corticosteroid treatment or require excessively high maintenance doses to control hemolysis [25]. Published relapse rates in wAIHA after first-line corticosteroid therapy vary widely, ranging from 10.5– 68 % among patients who initially respond to treatment [26–28]. In such cases, second-line treatments are employed, mainly splenectomy, rituximab, and immunosuppressants.

The present review aimed to evaluate and summarize the available evidence on adult patients with wAIHA in Latin America. Specifically, this study intended to assess the epidemiology, patient journey (diagnosis, treatments, and outcomes), clinical burden (morbidity and mortality), and healthcare resource utilization among patients with wAIHA, as well as to identify any unmet needs in the management of this condition.

This scoping review is reported in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses extension for Scoping Reviews (PRISMA-ScR) [29].

The present scoping review aimed to evaluate and summarize the available evidence reported on adult patients with wAIHA in Latin America. A thorough literature search was performed, encompassing various databases and gray literature. Therefore, we are confident that we have identified the majority of available literature on wAIHA in Latin America.

As is the case outside Latin America, studies on wAIHA are lacking and the scarce evidence available is primarily derived from retrospective single-center case series of patients.

As expected, the results show that women are disproportionally affected by wAIHA. This is likely explained by the fact that women have an increased susceptibility to autoimmune diseases [5], and is aligned with published literature [23,26]. The present review also confirms that systemic lupus erythematosus is a frequent underlying cause of wAIHA, being the most common in five of six studies that included patients with secondary wAIHA. Literature data show that AIHA occurs in 5–10 % of patients with systemic lupus erythematosus [37].

Corticosteroids are the mainstay first-line treatment for wAIHA, being recommended as such in both the Mexican guidelines and the Brazilian CPTG [17,36]. Accordingly, all but three of the 349 patients with non-missing information enrolled in the included studies received corticosteroids as the first-line treatment, in the great majority of cases as monotherapy (90.8 %). Prednisone, methylprednisolone, and dexamethasone were used.

Response rates to first-line treatments based on corticosteroids alone ranged from 68.6–84.3 % in these studies. Differences in response rates between studies might partly result from the variability in doses and durations of corticosteroid treatment and definitions of treatment response. These findings align with published literature reporting that between 70–80 % of patients generally respond to corticosteroid treatment [23,24]. The attainment of a CR to first-line corticosteroid treatment is particularly important, as this is associated with long-term remission. In a study by Roumier et al. [24], long-term remission was recorded in 92 % and 53 % of patients with a CR and a PR to first-line corticosteroids, respectively. Moreover, patients with PR may still present mild to moderate anemia, as the cut-off points defined for Hb in the response criteria were >9 g/dL and ≥10 g/dL in the three included studies that specified this information [5,14,30]. The CR rates observed in the studies included in this review (33.3–50.6 %) are also consistent with published literature [24,26]. The median relapse-free survival after first-line corticosteroid treatment differed significantly between the two studies that reported this outcome: 11.7 and 6.6 months in primary and secondary wAIHA [5], respectively, versus 81.7 months in a study where approximately half of the patients had secondary wAIHA [14]. This discrepancy may be, at least in part, a result of differences in relapse definitions, study populations, and type and duration of first-line corticosteroid treatment between the studies.

Though response rates to corticosteroid treatment were high, one should note that a considerable proportion of patients required second-line therapies (15.6–40.4 %). These findings converge with those reported by other studies, which showed that the rate of long-sustained remission after tapering and discontinuation of first-line corticosteroids is low (30–40 % after one year) [22,38,39]. Additionally, other literature found proportions of patients requiring second-line treatments after an initial response to corticosteroids (28.3 % and 52.2 %) similar to those in the present review [24,26]. Information on corticosteroid dependency was scarce in the included studies. Published data show that a high proportion of corticosteroid responders become corticosteroid dependent. In a study conducted in France, the rate of corticosteroid dependency among patients initially responding to corticosteroid treatment was above 60 % [24]. In another study of 44 patients receiving first-line prednisone, nine (20.5 %) required long-term maintenance immunosuppression. Of these nine individuals, seven also required the introduction of second-line agents [34].

While the first-line treatment of wAIHA was based almost exclusively on corticosteroids, the second line was characterized by various treatment options. Second-line treatment of wAIHA has classically been based on splenectomy, the effectiveness of which is similar to that of rituximab while providing slightly longer remissions [7]. Nevertheless, reports outside Latin America have described a gradual decline in the use of this invasive procedure, which is currently employed in <10 % of cases [7,40]. This decline has mostly been attributed to the resulting increase in infection risk and thrombotic complications. Moreover, splenectomy entails some additional downsides, namely the lack of reliable predictors of outcome and other potential surgical complications [7,40]. We found different practices regarding the use of splenectomy as the second-line treatment of wAIHA in Latin American centers. These practices ranged from a complete absence of splenectomies in two of the studies (in Chile [31] and Cuba [32]) to splenectomy being the sole second-line treatment in a referral center in Mexico City [5].

European data have demonstrated an emergence of rituximab as the preferred second-line option in the last decades [24]. Our findings show that rituximab was used in only two out of the 93 patients (2.2 %) who received second-line treatments in the included studies. Moreover, due to the limited number of studies and countries contributing data, these results may not accurately reflect the current use of rituximab in the management of wAIHA throughout Latin America. One of the possible reasons behind the limited use of this monoclonal antibody is the fact that rituximab is not currently approved for the treatment of wAIHA. Reasons reported by the retrieved studies included the unavailability of rituximab at some centers, as well as economic constraints limiting its use for the treatment of wAIHA [5,34].

The healthcare burden among relapsing patients is significant and underestimated [3]. These patients have a higher transfusion burden and a higher rate of alloimmunization [41]. Currently, the treatment of patients with relapsing disease and unfit for or refusing splenectomy constitutes an important clinical challenge [42,43]. Treatment options for such patients are few (namely azathioprine, cyclophosphamide, and cyclosporine) and of limited efficacy. Novel agents currently being studied in wAIHA, such as bortezomib, nipocalimab, rilzabrutinib, and ianalumab, might become important future treatment options for patients with relapsing disease [42–46].

The present literature review shows that several data gaps and challenges related to wAIHA currently exist in Latin America. First, epidemiological and healthcare resource utilization data are scarce, and no data were retrieved on patient-related outcomes, quality of life, or perspectives on wAIHA and its daily burden. In addition, whereas all seven studies evaluated the effectiveness of treatments given to wAIHA patients, only one reported safety results, specifically evaluating the short- to long-term safety of low-dose rituximab plus HDD as first-line therapy. Furthermore, data on blood transfusions among these patients were provided by only three of the studies and were limited to the proportion of those receiving transfusions during follow-up. Additional details, such as the number of transfusions per patient, patient response to transfusion, or the setting in which the transfusion was performed (inpatient versus outpatient), which would be essential to fully characterize transfusion burden in wAIHA patients, were not reported. Second, all studies determined response and/or remission rates as a means of evaluating the effectiveness of wAIHA treatments. However, other important treatment-related outcomes, such as time to response and relapse-free survival, have seldom been reported. Third, there are currently no standardized criteria to evaluate response (complete or partial) to wAIHA treatments, as is the case internationally [47,48]. The variability in the definitions used impairs the validity of cross-study comparisons. Lastly, information on patients refractory to wAIHA treatment, particularly to third line therapies and beyond, is scarce. However, the data reported by Jaime-Pérez suggest that an appreciable proportion of wAIHA patients (over 15 %) require more than two lines of treatment [14]. Further research is warranted to understand the management strategies and unmet needs of patients who fail to control the disease after several lines of treatment.

This study is not without limitations. Only four Latin American countries were represented in the present literature review, with studies conducted in Cuba accounting for 42.9 % of those included. As a result, our findings might not be representative of the reality throughout Latin America. In addition, most studies were small case series (four included <50 patients), and all were retrospective.

Though the population of interest in this review were adult patients with wAIHA, we decided to include studies that did not report data exclusively for patients ≥18 years or with the wAIHA subtype. Three of the studies, Quintero [30], Jaime-Pérez et al. [14], and Hernandez-Company et al. [5], presented data for samples of wAIHA patients aged ≥15, ≥16, and ≥17 years, respectively. In the study by Valdés et al. [33], two of the 35 enrolled patients had cold AIHA. The decision to not exclude these studies from the review was based on the scarcity of available data on wAIHA patients in Latin America. The impact of this limitation is expected to be minimal, as the study samples were still comprised almost exclusively of adult patients with wAIHA.

As previously shown, the definitions of CR and PR varied. Moreover, the timing of the assessment of treatment response differed between both studies that provided this information: three and four weeks after the start of first-line treatment with corticosteroids in the studies by Hernandez-Company et al. [5] and Agramonte et al. [32], respectively. The full extent of this limitation is difficult to ascertain as information on the timing of treatment response assessment was not available in the remaining studies. Both these factors might influence response rates (e.g., more stringent response criteria and earlier assessment of response to treatment may both lead to lower response rates) and should thus be taken into account when comparing the results of these studies.

The treatment outcome rates (response or remission) reported by the individual studies, particularly for treatments beyond the first line, should be interpreted cautiously considering the small number of treated patients.

This work was supported by Johnson & Johnson Innovative Medicine (LATAM region). The sponsor was involved in the study design, collection and interpretation of data, writing of the report, and in the decision to submit the article for publication.

Yudy Medina, Renato Watanabe de Oliveira, and Pamella Villanova developed the protocol outline and search strategy. Yudy Medina and Renato Watanabe de Oliveira conducted the literature search, which was reviewed by Diogo Ribeiro. Yudy Medina and Diogo Ribeiro performed the data extraction. Diogo Ribeiro drafted the manuscript draft, and all other co-authors contributed to writing, reviewing, and editing. All authors read and approved the final manuscript.

The data that support the findings of this study are available from the corresponding author upon reasonable request.