Adult aplastic anemia in Thailand: incidence and treatment outcome from a prospective nationwide population-based study

The incidence and outcomes of aplastic anemia (AA) in Asia remain limited. This study aimed to explore the incidence and outcomes of patients with adult AA across the country of Thailand. This is a prospective multi-center nationwide population-based observational study of AA patients aged at least 15 years old, diagnosed from August 2014 to July 2016, with a longitudinal follow-up period over 2 years. There were 348 newly diagnosed adult AA patients during the enrollment period, giving an annual incidence of 4.6 per million. The incidence of severe (SAA) and very severe aplastic anemia (VSAA) (3.8 per million) was higher than non-severe AA (NSAA, 0.8 per million). The peak incidence was observed in the patients aged from 80 to 89 years old (14.4 per million). The 2-year overall survival (OS) in NSAA, SAA, and VSAA were 65.5%, 49.3%, and 20.1%, respectively (P < 0.001). With regard to the response to immunosuppressive therapy, the overall response rate (ORR) in SAA/VSAA treated with rabbit anti-thymocyte globulin with/without cyclosporin A (rATG ± CsA) were significantly superior to those treated with CsA alone, or anabolic steroids (44.4% vs 36.4% and 31.2%, respectively, P < 0.001). The 2-year OS in SAA/VSAA treated with rATG ± CsA, CsA, and anabolic steroids were 54.8%, 54.5%, and 37.6% (P = 0.037), respectively. The incidence of adult AA in Thailand is higher than those in Western countries, and the peak incidence is in the elderly. rATG ± CsA provided a better response than anabolic steroids, translating to the superior survival in SAA/VSAA treated with rATG ± CsA. Supplementary Information The online version contains supplementary material available at 10.1007/s00277-021-04566-0.

Immunosuppressive therapy (IST) is recommended for the treatment of SAA patients who are not eligible for hematopoietic stem cell transplantation (HSCT) [12,13]. Although horse ATG (hATG) gives better responses over rabbit ATG (rATG) as first-line therapy [13], rATG is widely used as initial therapy in Europe and Asia due to unavailability of the hATG. Recent multi-national studies of rATG in Europe and Asia showed 1-year response rates of 63-65% with favorable survival outcomes (2-year overall survival 79.0-86.3%) [14][15][16]. Anabolic steroids were used as an alternative therapy for those unaffordable or unsuitable for HSCT or IST. To understand the current situation of adult AA in Thailand, the annual incidence of AA across the country and the treatment responses as well as survival outcome according to patient's characteristics and treatment modalities were studied.

Study design and patients
This study was a multi-center prospective observational study. Newly diagnosed adult AA patients from 30 referral centers nationwide (aged at least 15 years) were enrolled during the period from August 1, 2014, to July 31, 2016. Criteria for diagnosis of AA included bone marrow hypocellularity with at least two of the following: anemia with corrected reticulocyte count ≤ 1%, absolute neutrophil count (ANC) ≤ 1.5 × 10 9 /L, or platelet count ≤ 50 × 10 9 /L. Patients with myelodysplastic syndrome (MDS) or classic paroxysmal nocturnal hemoglobinuria and chemotherapyinduced bone marrow suppression were excluded. Patients with chromosomal abnormalities were also excluded. Those with nationalities other than Thai were excluded. All enrolled patients were reassessed for definite diagnosis by the central review committee. Treatment of choice was given at the discretion of the treating physicians, depending on the patient's eligibility and accessibility to treatment. The research proposal and all subsequent amendments were approved by the local Ethics Committee at each study site. The study was conducted in accordance with ethical standards of the responsible committee on human experimentation and with the Helsinki Declaration of 1975, as revised in 2008. All patients provided written informed consent.

Responses
Responses to treatment were evaluated at each visit every 3 months. In those who switched to subsequent therapy, the responses were reassessed to evaluate the efficacy of the second-line treatment. Criteria for responses were based on the British Committee for Standard Haematology (BCSH) guideline [12,17]. Briefly, the response criteria were as follows: complete response (CR): normal hemoglobin (Hb) for age and gender with ANC ≥ 1.5 × 10 9 /L and platelet ≥ 150 × 10 9 /L; partial response (PR): transfusion independence and no longer met the criteria for severe disease, and no response (NR): not achieving CR/PR. The response criteria for NSAA were similar to SAA for CR and NR, whereas the criteria for PR were transfusion independence (if previously dependent) or doubling or normalization of at least one lineage or increase of baseline Hb ≥ 3 g/dL, ANC ≥ 0.5 × 10 9 /L (if initially < 0.5 × 10 9 /L), and platelet ≥ 20 × 10 9 /L (if initially < 20 × 10 9 /L).

Statistical analyses
The incidence rate was calculated by the number of newly diagnosed AA patients divided by the number of inhabitants within the respective catchment areas. All eligible cases were recruited from almost all hospitals over the country. We excluded hospitals in which there were no hematologists. The catchment areas were defined to only the provinces with the hematology services ( Fig. 1). Only patients who had lived at least 6 months in the catchment area were included into the incidence calculation. Population numbers were collected from the national database provided by the Strategy and Planning Division, Office of Permanent Secretary, Ministry of Public Health (http:// bps. moph. go. th/ new_ bps/). Direct age standardization was presented as previously described [18].
Comparisons of data were performed using t-test or Wilcoxon rank-sum test or χ 2 , depending on variable types and data distribution. OS was analyzed using Kaplan-Meier plots. Cox regressions were used to estimate hazard ratios among given subgroup stratifications. Treatment responses to immunosuppressive therapy were pre-defined as previously described criteria [17]. The statistical analyses were performed using SAS 9.4.

Results
A total of 348 patients were enrolled. Twenty-six patients were excluded for the incidence calculation because they had lived in the catchment area less than 6 months. Six patients were excluded from the treatment response analyses due to inadequate collected data.

Incidence of AA
There were 322 patients with newly diagnosed adult AA during the 2-year periods, which yielded an overall annual incidence of 4.6 per million. The annual incidence of SAA/VSAA was 3.8 per million, higher than that of NSAA (0.8 per million with a ratio of 4.6:1). Forty-five patients had VSAA, giving the annual incidence of 0.6 per million. The incidence was high in older patients with the peak of 14.4 per million in patients aged from 80 to 89 years old (Table 1). It was lower in patients aged less than 50 (1.2-2.3 per million). There was a slightly higher incidence in men than women (4.8 vs 4.0 per million). Age-standardized incidence rates (ASRs) were reported by geographical regions, which showed the highest incidences in eastern and north-eastern areas with ASRs of 6.6 and 5.6 per million, respectively. Variation in incidence rate was observed across the country, ranging from 2.6 to 6.6 per million ( Fig. 1, Table S1).
One hundred thirty-seven (137) patients with SAA/VSAA did not receive frontline rATG. The reasons included patient conditions not suitable for IST (such as active infections and comorbidities, n = 58), inaccessibility to rATG (n = 35), concerns regarding medical facility (n = 17), and other reasons (n = 27).
In 59 NSAA patients, the majority of them (n = 49, 83.1%) received anabolic steroids as frontline therapy, while 8 received supportive treatment. The other 2 patients received CsA.

Discussion
In this nationwide prospective study of adult AA patients diagnosed from 2014 to 2016, the annual incidence of adult AA in Thailand was 4.6 per million, which considerably is more or less similar to 3.0-5.0 per million in the previous population-based study in Thailand conducted 20 years ago [6]. Similarity in the annual incidence is also noted among regional parts: 2.8 per million in the southern region (vs 3.0 per million in Songkla and nearby provinces) and 5.6 per million in the north-eastern region (vs 5.0 per million in Khon Kaen and nearby provinces) [6]. The steadiness of the incidence through decades, even the changes in lifestyle and environment toward urbanization, favors the genetic predisposition of autoimmunity as the disease etiology [19,20]. Nevertheless, the environmental exposer may have an impact on the variation of geographical incidence as the lowest was in the central region (2.5 per million), whereas the incidence was higher in the north-eastern region, the largest agricultural area (5.6 per million) and the peak was in the eastern region, the major industrial district (6.5 per million). The AA annual incidence of 4.6 per million being reported here is approximately twice higher than that from Western countries (ranged 1.5-2.3) [1][2][3][4][5] but is similar to the reports from Asia: 4.8 in Malaysia [21], 5.7 in Taiwan [9], and 7.5 in China [8,22]. The current study demonstrated a remarkable agerelated distribution with steeply increased incidence after the age of 50 (6.2 per million) with a peak incidence at the age over 80 years old (14.3 per million). The pre-existing clonal hematopoiesis, causing bone marrow failure with overlapping features of hypocellular MDS and AA, may lead to a higher incidence of AA among elderly patients [23,24]. Due to a high proportion of patients with SAA/ VSAA in this study (83%) and half of them older than 60 years, the 2-year OS was only 48%, compared with the other reports from Sweden (5-year OS 90.7%) [5], Taiwan (5-year OS 60%) [9], and Spain (2-year OS 57%) [3]. Following the previous population-based studies [3,5,9], older age and severity of the disease were the strong predictors of inferior survival among patients with AA.
HSCT is recommended treatment for younger patients of less than 40-50 years old. However, not all eligible patients could undergo this expensive procedure. Furthermore, the government supports only specific number of patients. Therefore, only small number of patients underwent HSCT in this cohort.
rATG is the mainstay of treatment for SAA in Thailand as well as in many countries in Asia. For patients with SAA/VSAA in this cohort, the ORR was considerably low (44.4%), compared with previously published reports of rATG [14,16]. However, when considering the response among the evaluable patients, the response gradually increased over time with the ORR of 44% at 6 months and 68% at 1 year, consistent with the reports of rATG from Asia (ORR 17% at 6 months and 64% at 12 months) and European Blood and Marrow Transplant (EBMT) (ORR 52% at 6 months and 64% at12 months) [14,16]. Although the response to rATG in this study was delayed, it was increased over time and the favorable response of rATG was a surrogate marker of long-term survival, similar to previous reports [14,25].
Anabolic steroids have long been used to treat AA. Its mechanisms include increasing telomerase (TERT) gene expression in hematopoietic cells [26], stimulating erythroid progenitor [27], and enhancing regulatory T-cell [28]. The responses to anabolic steroids are varied, ranging from 25 to 56%, depending on the severity of AA and the treatment combination [29][30][31][32]. In this cohort, the response to anabolic steroids appeared to improve over time with an acceptable response in patients with NSAA (40.8%). In SAA/VSAA, the responses were unfavorable (31.2%) compared with rATG ± CsA (44.4%). Giving the inferior OS among patients treated with anabolic steroids, anabolic steroids should therefore be reserved for patients with NSAA and those with SAA/VSAA who are unfit for IST and HSCT.
The outcomes of this study represented the real-world situation of patients with AA. The cases were collected from 30 hospitals, including university and public hospitals, located in all regions of the country; therefore, the data would reasonably reflect the actual AA incidence in Thailand. The further strengths of this study included the prospective design and centrally reviewed diagnosis. However, this study had some limitations. First, the incidence of NSAA may be underestimated especially non-transfusion dependent AA and asymptomatic cases. Second, there was a proportion of non-evaluable patients due to death and loss to follow-up. Third, this study was not randomized, and the comparison between treatment modalities should therefore be interpreted with caution as there would be a high selection bias for older and unfit patients to receive anabolic steroids or supportive care.
In conclusion, the incidence of adult AA in Thailand from this nationwide population-based study is higher than in Western countries, and even higher among the elderly. The real-world outcome of patients with SAA, especially in those aged over 60 years, is substantially poor. The appropriate therapeutic options, as well as the accessibility to advanced treatment, are needed to ensure a better outcome of patients.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http:// creat iveco mmons. org/ licen ses/ by/4. 0/.