Genotypic distribution and molecular spectrum of rare and novel thalassemia variants in Ganzhou, southern China Authors: Jungao Huang, Xiaoyan Huang, Xinxing Xie, Xiaoqin Xin Published: 2026 ABSTRACT BackgroundThalassemia is the most common single-gene inherited blood disorder worldwide, being relatively rare in northern China but more prevalent in southern China. To facilitate effective prevention and control of thalassemia, we analyze the genotype and frequency distribution of rare thalassemia variants in the general population of the Ganzhou area, offering valuable insights for genetic counseling and prenatal diagnosis.MethodsBetween January 2022 and January 2025, a cohort of 84,067 individuals was screened in Ganzhou. Following the exclusion criteria, 1,081 participants with rare thalassemia variants were included in the final analysis. Genotypes associated with thalassemia were determined using next-generation sequencing (NGS). Variant classification adhered to ACMG/AMP guidelines.ResultsAmong the participants with rare thalassemia variants, 495 individuals presented with rare α-thalassemia variants and 560 with rare β-thalassemia variants, including 26 newly identified variants. Rare thalassemia variants accounted for 1.29% of the total screened population. The distribution of rare thalassemia variants exhibited regional variation across Ganzhou, with the highest rates in Nan Kang (12.69%), followed by Xingguo (10.49%) and Zhanggong (8.80%). Chongyi had the lowest rate at 1.78%, with all these locations located to the northwest of Ganzhou. Additionally, 26 individuals carried novel thalassemia variants. Six of these variants were classified as “Likely Pathogenic,” and the HBB:c.50G>T (Gly > Val) genotype was associated with hypochromic anemia.ConclusionThis study offers a comprehensive analysis of rare and newly identified thalassemia variants in the Ganzhou area, highlighting the complexity and heterogeneity of the condition. These findings underscore the necessity for effective screening methods in regions with a high incidence of thalassemia and provide essential insights for the targeted prevention and management of this disorder in the future. Introduction Thalassemia is a genetic hematological disorder characterized by insufficient hemoglobin production, affecting millions of individuals worldwide. Approximately 5.2% of the global population is afflicted, making it one of the most prevalent single-gene diseases (Tarim and Oz, 2022). Each year, between 300,000 and 400,000 infants are born with various forms of thalassemia, underscoring the significant global health burden associated with this condition (Vachhani et al., 2022). In China, the prevalence of thalassemia is particularly high in the southern regions, especially south of the Yangtze River, where carrier rates rank among the highest in the country (Wang WD. et al., 2023). The Ganzhou area, known for its elevated incidence of thalassemia (Zhao et al., 2018), is home to a significant population of Hakka people, exhibiting an alarming carrier rate of 14.54% (Xie et al., 2023). Clinically, thalassemia primarily manifests as hypochromic microcytic hemolytic anemia and is often accompanied by symptoms such as splenomegaly, jaundice, and characteristic skeletal deformities (Ansharullah et al., 2025). Thalassemia can be categorized into several types, including α-, β-, δ-, and γ-thalassemia, with α- and β-thalassemia being the two predominant forms encountered in clinical practice. These conditions arise from the loss or mutation of the β-globin gene (HBB) or the α-globin genes (HBA1 and HBA2), respectively (Zhang H. et al., 2025). The severity of clinical manifestations and anemia varies based on the specific genetic variant. While thalassemia minor may present as mild anemia or remain asymptomatic, thalassemia major typically results in severe anemia (Zakaria et al., 2021). Despite advancements in therapies such as blood transfusions, bone marrow transplants, and gene therapy, prevention remains the most effective and straightforward strategy (Hu et al., 2024; Laurent et al., 2024; Philippidis, 2024). Current preventive measures primarily utilize a three-tier screening approach, encompassing routine blood tests, hemoglobin electrophoresis, and genetic confirmation (Fakher et al., 2007; Li et al., 2023). However, conventional testing mainly targets 23 common variants—three mutation types and three deletions for α-thalassemia and 17 mutations for β-thalassemia—thus leaving many rare variants undetected and complicating prevention efforts. To date, over 1,000 thalassemia variants have been documented in the specialized globin gene server database, HbVar (https://globin.bx.psu). Notably, these conventional targeted approaches continue to be widely used in routine screening and diagnostic settings, owing to their established reliability, low bioinformatics demands, and straightforward scalability for population-level screening. Recent advancements in genomic technologies, particularly next-generation sequencing (NGS), have improved the detection of rare thalassemia variants. Previous studies indicate that direct NGS could reduce misdiagnosis rates by 30.5%–74.8% compared to traditional genetic screening methods (He et al., 2017; Yang et al., 2023; Cao et al., 2026). However, many earlier studies did not compile a comprehensive catalog of rare thalassemia cases, and it remains uncertain whether the prevalence of these rare variants varies across different regions. To enhance thalassemia prevention efforts, we employed NGS techniques in the Ganzhou area to identify rare and newly emerging thalassemia variants. This initiative aims to inform and improve future governmental health policies. Our study involves a substantial cohort of 84,067 participants, spanning from January 2022 to January 2025, laying a robust theoretical foundation for thalassemia screening, targeted prevention strategies, and effective interventions. Participants and methods This study involved a cohort of 84,067 individuals screened in Ganzhou, Southern China, from January 2022 to January 2025. After excluding non-thalassemia carriers (n = 73,230), 10,837 participants with thalassemia variants remained. Further exclusion of those with variants detectable by traditional methods (n = 9,756) narrowed the final analysis to 1,081 participants with rare variants (Figure 1A). The majority of participants were local residents from various districts of Ganzhou, predominantly Han (99.1%), with a minority representing Hui and She ethnic groups (0.9%). Age distribution ranged from 1 to 87 years, with a mean age of 27 years. All individuals provided written informed consent prior to participation. Any variant identified by our NGS assay that fell outside the routine 23-variant panel was classified as “rare”. Variants that have not been previously described were classified as “novel”. The population in the study. (A) Flow chart of the enrollment of participants. (B) Constituent ratios of different age groups of 1081 participants carried rare/novel thalassemia variants. Blood samples were collected from participants and analyzed using an automated blood analyzer (UniCel DxH 800; Beckman, United States) to assess key hematological parameters, including hemoglobin concentration, mean corpuscular volume (MCV), and mean corpuscular hemoglobin (MCH). Participants were categorized as positive if MCH was <27 pg and/or MCV was <82 fL. Genomic DNA was extracted from peripheral blood samples using the QIAamp Blood Mini Kit (Qiagen; Hilden, Germany), following the manufacturer’s protocol. The purity and concentration of the extracted DNA were measured using a spectrophotometer (NanoDrop 2000; Thermo Scientific, United States), with only samples having concentrations >20 ng/mL and an A260/A280 ratio between 1.8 and 2.0 accepted for analysis. The sequencing method used was consistent with our previous reports (Xie et al., 2023). Briefly, full-length amplification of the HBA1, HBA2, and HBB genes was performed via polymerase chain reaction (PCR), ensuring coverage of all relevant exons and introns identified in the HbVar database (https://globin.bx.psu). The primer panel comprised six pairs of oligonucleotides, targeting four genes (HBA1, HBA2, HBB-1, and HBB-2) and two common deletion breakpoints (HBA-Q and HBB-Q). The PCR amplification was carried out in 96-well plate format, with each well corresponding to an individual sample library. A set of 96 unique index sequences was designed and assigned to each well, and the six target-specific primers labeled with these index sequences were designated as index primers, enabling sample barcoding and multiplex sequencing in a single run. The primers are associated with patents WO/2014/023076, WO/2014/023167, and CN102952877. NGS libraries were prepared according to the Illumina HiSeq Sequencing Library Preparation Protocol, and sequencing was conducted on the Illumina HiSeq2000 platform using paired-end sequencing for 100 base pairs. Subsequent bioinformatics analyses included variant calling and annotation, employing established methodologies to ensure accurate identification of genetic variants. Variants identified by NGS were validated by Sanger sequencing, which ensured the accuracy and reliability of all reported rare and novel variants. Common deletions associated with α-thalassemia or β-thalassemia were further validated using Gap-PCR (BGI, Shenzhen, China) and 4% agarose gel electrophoresis. Variant interpretation and classification followed the gene-specific ACMG/AMP guidelines established by the ClinGen Hemoglobinopathy Variant Curation Expert Panel (Kountouris et al., 2022). Statistical analysis All statistical analyses were performed using SPSS version 27.0 software (SPSS Inc., Chicago, IL, United States). Categorical variables are reported as counts (n) and percentages (%). Chi-square tests were employed to compare the detection rates of α-thalassemia and β-thalassemia. A two-sided P-value of <0.05 was considered statistically significant. Results A total of 1,081 participants with rare thalassemia variants were included in the final analysis. Among these, 495 individuals exhibited rare α-thalassemia variants, while 560 carried rare β-thalassemia variants, including 26 individuals with newly identified variants. Overall, rare thalassemia variants accounted for 1.29% of the total screened population (1,081 out of 84,067) (Figure 1A). In terms of age distribution, the 20–30 age group comprised 56.6% of participants, followed by 36.38% in the 30–40 age bracket, indicating a predominance of individuals of reproductive age (Figure 1B). The distribution of rare thalassemia variants displayed regional variation across Ganzhou, with the highest rates observed in Nan Kang (12.69%), followed by Xingguo (10.49%) and Zhanggong (8.80%). Chongyi recorded the lowest rate at 1.78%. Notably, all these locations are situated to the northwest of Ganzhou (Figure 2). Geographic distribution of rare thalassemia in Southern Jiangxi, China. Clinically, α/β thalassemia variants are categorized as severe, and mild (α0, and α+; β0, and β+). Among the 495 individuals with rare α-thalassemia variants, we identified 41 distinct types, with Hb Hekinan II (HBA1:c.84G>T) (α+) being the most prevalent at 28.283%, corresponding to a population frequency of 0.167% (140 out of 84,067). The αα/--THAI (α0) variant constituted 3.030%. The top nine genotypes of rare α-thalassemia accounted for the majority, representing 86.06% of the cases (Table 1). Among the 560 individuals with rare β-thalassemia, we identified 44 types, with the 32019UTR +129 (T>A) (β+) variant being the most common at 27.679%, translating to a frequency of 0.184%. The β0 variant with the highest occurrence was (SEA)-HPFH, observed at a rate of 2.857%. Similarly, the top nine genotypes of rare β-thalassemia represented the majority, comprising 83.93% of the total (Table 2). TABLE 141 rare variants of α-thalassemia, and their frequency in the population.CategoryRare variantsHGVS (Nucleotide)HGVS (Protein)ClassNumber(n)Ratio (%)Frequency (%)1Hb Hekinan IINM_000558.5:c.84G>Tp.Trp28Cysα+14028.283%0.167%2Hb Q-ThailandNM_000558.5:c.223G>Cp.Asp75HisHb variant7314.747%0.087%3Hb OwariNM_000558.5:c.364G>Ap.Asp122AsnHb variant459.091%0.054%4Hb G-HonoluluNM_000517.6:c.91G>Cp.Asp31HisHb variant438.687%0.051%5Hb NanchangNM_000517.6:c.46G>Ap.Gly16SerHb variant326.465%0.038%6Hb BinyangNM_000517.6:c.40G>Tp.Ala14SerHb variant285.657%0.033%7HKααNG_000006.1:g.?_?—α+265.253%0.031%8Hb Phnom PenhNG_000006.1:g.?_?—α+244.848%0.029%9αα/--THAING_000006.1:g.10664_44164del33501—α0153.030%0.018%10Initiation codon (ATG>A-G)NM_000558.5:c.2T>Gp.Met1Argα+132.626%0.015%11Hb Hengqin INM_000558.5:c.16G>Ap.Gly6SerHb variant61.212%0.007%12Hb LiaobuNM_000517.6:c.322G>Cp.Asp108HisHb variant40.808%0.005%13Hb HollythorpeNM_000517.6:c.96-12T>ANon-coding (intronic)Hb variant40.808%0.005%14Hb VenetiaNM_000517.6:c.65C>Tp.Pro22LeuHb variant30.606%0.004%15Hb Ube-2NM_000558.5:c.205A>Gp.Asn69AspHb variant30.606%0.004%16Hb OttawaNM_000517.6:c.46G>Cp.Gly16ArgHb variant30.606%0.004%17Alpha2 Codon 30 del GAGNM_000517.6:c.91_93delGAGp.Glu31delα+30.606%0.004%18α2 Codon 31 (AGG>AAG)NM_000517.6:c.94G>Ap.Glu32Lysα+20.404%0.002%19Codon 61 (AAG>TAG) [Lys > STOP]NM_000558.5:c.184A>Tp.Lys62Terα+20.404%0.002%20Codon 22 (GGC>GGT)NM_000517.6:c.67G>Tp.Gly23Cysα+20.404%0.002%21Hb Zurich-AlbisriedenNM_000558.5:c.178G>Cp.Ala60Proα+20.404%0.002%22Hb Port PhillipNM_000517.6:c.275T>Cp.Leu92ProHb variant20.404%0.002%23Hb Groene HartNM_000558.5:c.358C>Tp.Pro120Serα+20.404%0.002%24IVS-I-38 (C>T)NM_000517.6:c.95-38C>TNon-coding (intronic)α+10.202%0.001%25IVS-I-116 (A>G)NM_000517.6:c.95-116A>GNon-coding (intronic)α+10.202%0.001%26IVS-I-1 (AGGT>AGAT) donorNM_000517.6:c.95+1G>ANon-coding (intronic)α+10.202%0.001%27Hb XuchangNM_000517.6:c.133C>Tp.Pro45SerHb variant10.202%0.001%28Hb St. Luke’sNM_000558.5:c.287C>Gp.Pro96ArgHb variant10.202%0.001%29Hb NorthwoodNM_000558.5:c.293A>Gp.Asn98SerHb variant10.202%0.001%30Hb MacarenaNM_000517.6:c.358C>Tp.Pro120SerHb variant10.202%0.001%31Hb LuxembourgNM_000558.5:c.73T>Cp.Trp25ArgHb variant10.202%0.001%32Hb Liuzhou-YufengNM_000558.5:c.334G>Tp.Asp112TyrHb variant10.202%0.001%33Hb J-TorontoNM_000558.5:c.17C>Ap.Ala6AspHb variant10.202%0.001%34Hb J-TashikuerganNM_000558.5:c.59C>Ap.Ala20AspHb variant10.202%0.001%35Hb Garden stateNM_000558.5:c.248C>Ap.Pro83GLnHb variant10.202%0.001%36Hb DagestanNM_000558.5:c.181A>Gp.Lys61GLuHb variant10.202%0.001%37Hb BroomhillNM_000558.5:c.343C>Gp.Pro115AlaHb variant10.202%0.001%38Fusion gene——α+10.202%0.001%39Codons 90–92 (-8bp): (-AGCTTCGG)NM_000558.5:c.270_277delAGCTTCGGp.?α+10.202%0.001%40Hb Amsterdam-A1NM_000558.5:c.99G>Ap.Trp33Terα+10.202%0.001%41Hb DapuNM_000517.6:c.52G>Tp.Ala18SerHb variant10.202%0.001%Total495100.000%0.589% 41 rare variants of α-thalassemia, and their frequency in the population. TABLE 244 rare variants of β-thalassemia, and their frequency in the population.CategoryRare variantsHGVS (Nucleotide)HGVS (Protein)ClassNumber(n)Ratio (%) Frequency (%) 13′UTR +129 (T>A)NM_000518.5:c.*129T>ANon-coding (3′UTR)β+15527.679%0.184%2−50 (G>A)NM_000518.5:c.-50G>ANon-coding (5′UTR)β+7112.679%0.084%3Hb New YorkNM_000518.5:c.341T>Ap.Val114GLuHb variant5710.179%0.068%4Chinese Gγ+(Aγδβ)0NG_000007.3:g.48795_127698del78904—β0478.393%0.056%5Hb ZengchengNM_000518.5:c.343C>Ap.Pro115ThrHb variant468.214%0.055%6Hb J-BangkokNM_000518.5:c.170G>Ap.Gly57AspHb variant396.964%0.046%7−63 (A>G)NM_000518.5:c.-63A>GNon-coding (5′UTR)β+305.357%0.036%8(SEA)-HPFHNG_000007.3:g.48795_127698del78904—β0162.857%0.019%9−90 (C>T)NM_000518.5:c.-90C>TNon-coding (5′UTR)β+91.607%0.011%10Hb Zurich-LangstrasseNM_000518.5:c.151A>Tp.Met51LeuHb variant81.429%0.010%11Codons 54–58 (-TATGGGCAACCCT)NM_000518.5:c.165_177delTATGGGCAACCCTp.Leu19_Leu20del?β071.250%0.008%12Taiwanese deletionNG_000007.3:g.48795_127698del78904—β061.071%0.007%13CAP +8 (C>T)NM_000518.5:c.-8C>TNon-coding (5′UTR)β+61.071%0.007%14−72 (T>A)NM_000518.5:c.-72T>ANon-coding (5′UTR)β+61.071%0.007%15−42 (-C)NM_000518.5:c.-42delCNon-coding (5′UTR)β+61.071%0.007%165′UTR +22 (G>A)NM_000518.5:c.-15G>ANon-coding (5′UTR)β+50.893%0.006%17Hb MolfettaNM_000518.5:c.379G>Cp.Asp127HisHb variant50.893%0.006%18Hb J-TaichungNM_000518.5:c.389C>Ap.Ala130AspHb variant50.893%0.006%19Hb G-SirirajNM_000518.5:c.22G>Ap.Gly8SerHb variant40.714%0.005%20Hb HamiltonNM_000518.5:c.34G>Ap.Gly12ArgHb variant30.536%0.004%21Hb G-TaipeiNM_000518.5:c.68A>Gp.Asp23GLyHb variant30.536%0.004%22Hb Taipei-TienNM_000518.5:c.122G>Tp.Gly41ValHb variant30.536%0.004%23Hb RenertNM_000518.5:c.401T>Cp.Val134AlaHb variant20.357%0.002%24Cap +1570 (T>C)NM_000518.5:c.*1570T>CNon-codingβ+10.179%0.001%255′UTR +28 (A>G)NM_000518.5:c.-9A>GNon-coding (5′UTR)β+10.179%0.001%263′UTR +132 (C>T)NM_000518.5:c.*132C>TNon-coding (3′UTR)β+10.179%0.001%273′UTR +118 (A>G)NM_000518.5:c.*118A>GNon-coding (3′UTR)β+10.179%0.001%28−32 (C>T)NM_000518.5:c.-32C>TNon-coding (5′UTR)β+10.179%0.001%29Hb DhonburiNM_000518.5:c.380T>Gp.Leu127Argβ+10.179%0.001%30Hb Cochin-Port RoyalNM_000518.5:c.440A>Gp.Asp147GLyHb variant10.179%0.001%31IVS-I (−2) or codon 30 (A>G)NM_000518.5:c.30A>GNon-coding (intronic)β010.179%0.001%32Poly A (AATAAA>AATGAA)NM_000518.5:c.*110T>CNon-coding (3′UTR)β+10.179%0.001%33IVS-II-63 (T>C)NM_000518.5:c.316-63T>CNon-coding (intronic)β+10.179%0.001%34IVS-II-613 (C>T)NM_000518.5:c.316-613C>TNon-coding (intronic)β+10.179%0.001%35IVS-I-129 (A>G)NM_000518.5:c.129A>GNon-coding (intronic)β010.179%0.001%36Hb Bologna-St.OrsolaNM_000518.5:c.439C>Tp.Pro147SerHb variant10.179%0.001%37Hb AlpertonNM_000518.5:c.407C>Tp.Pro136LeuHb variant10.179%0.001%38Hb City of HopeNM_000518.5:c.208G>Ap.Gly70SerHb variant10.179%0.001%39Hb KofuNM_000518.5:c.254C>Tp.Pro85LeuHb variant10.179%0.001%40Hb SitiaNM_000518.5:c.386C>Tp.Pro129LeuHb variant10.179%0.001%41Hb RambamNM_000518.5:c.209G>Ap.Gly70AspHb variant10.179%0.001%42Hb MidnaporeNM_000518.5:c.161C>Tp.Pro54LeuHb variant10.179%0.001%43Hb Pierre-BéniteNM_000518.5:c.273G>Cp.Lys91AsnHb variant10.179%0.001%44Hb OlympiaNM_000518.5:c.61G>Ap.Gly21SerHb variant10.179%0.001%Total560100.000%0.666% 44 rare variants of β-thalassemia, and their frequency in the population. Additionally, 26 individuals exhibited novel thalassemia variants, totaling 23 distinct types, all in a heterozygous state. Six of these variants were classified as “Likely Pathogenic”: HBA1:c.223G>T (p.Asp75Tyr), HBA1:c.268C>G (p.His90Asp), HBA2:c.63C>G (p.His21GLn), HBB:c.-159delA (non-coding, 5‘UTR), HBB:c.316-3C>T (non-coding, intronic), and HBB:c.50G>T (p.Gly17Val). The remaining variants were categorized as “Variants of Uncertain Significance” (VUS) (Table 3). Hematological evaluations revealed that individuals with the HBB:c.50G>T (p.Gly17Val) genotype presented with hypochromic anemia, whereas the clinical phenotypes of the other genotypes did not exhibit significant alterations (Table 4). TABLE 3The distrubition of 23 novel variants in thalassemia.CategoryNovel variantsAmino acid change (HGVS)StatusNumber of casesMaleFemalePathogenity1HBA1:c.*13_14insTNon-coding (3‘ UTR)Heterozygous11-VUS2HBA1:c.*35delGNon-coding (3‘ UTR)Heterozygous1-1VUS3HBA1:c.*57delCNon-coding (3‘ UTR)Heterozygous1-1VUS4HBA1:c.*74_75insCNon-coding (3‘ UTR)Heterozygous2-2VUS5HBA1:c.*94A>CNon-coding (3‘ UTR)Heterozygous1-1VUS6HBA1:c.*95A>TNon-coding (3‘ UTR)Heterozygous11-VUS7HBA1:c.202A>Gp.Thr68AlaHeterozygous1-1VUS8HBA1:c.223G>Tp.Asp75TyrHeterozygous1-1Likely pathogenic9HBA1:c.268C>Gp.His90AspHeterozygous1-1Likely pathogenic10HBA2:c.*91_92insANon-coding (3’ UTR)Heterozygous1-1VUS11HBA2:c.118A>Gp.Thr40AlaHeterozygous1-1VUS12HBA2:c.300 + 43delGNon-coding (intronic)Heterozygous11-VUS13HBA2:c.301-35delGNon-coding (intronic)Heterozygous211VUS14HBA2:c.63C>Gp.His21GLnHeterozygous11-Likely pathogenic15HBA2:c.95 + 11delCNon-coding (intronic)Heterozygous1-1VUS16HBA2:c.95+9C>GNon-coding (intronic)Heterozygous1-1VUS17HBB:c.*145delANon-coding (3‘ UTR)Heterozygous1-1VUS18HBB:c.-159delANon-coding (5’ UTR)Heterozygous1-1Likely pathogenic19HBB:c.-28T>CNon-coding (5′ UTR)Heterozygous1-1VUS20HBB:c.316-3C>TNon-coding (intronic)Heterozygous2-2Likely pathogenic21HBB:c.-49C>TNon-coding (5′ UTR)Heterozygous1-1VUS22HBB:c.50G>Tp.Gly17ValHeterozygous1-1Likely pathogenic23HBB:c.93-53_93-52insTTCNon-coding (intronic)Heterozygous1-1VUSTotal26521VUS: variant of uncertain signifcance; -: NA, Variant annotations were performed using the following reference transcript sequences: HBA1: NM_000558.5, HBA2: NM_000517.6, and HBB: NM_000518.5. The distrubition of 23 novel variants in thalassemia. VUS: variant of uncertain signifcance; -: NA, Variant annotations were performed using the following reference transcript sequences: HBA1: NM_000558.5, HBA2: NM_000517.6, and HBB: NM_000518.5. TABLE 4The phenotypes of 6 novel variants which predicted to be pathogenic.Novel variantsAmino acid change (HGVS)GenderAge (years)RBC (1012/L) MCV(fl)MCH(pg)Hb(g/dL)HBA1:c.223G>Tp.Asp75TyrFemale424.4985.529.8134HBA1:c.268C>Gp.His90AspFemale305.0487.928.3143HBA2:c.63C>Gp.His21GLnMale265.1882.329.8155HBB:c.-159delANon-coding (5′ UTR)Female304.7185.829.4138HBB:c.316-3C>TNon-coding (intronic)Female313.7488.429.4110HBB:c.316-3C>TNon-coding (intronic)Female244.1794.128.9121HBB:c.50G>Tp.Gly17ValFemale245.3166.619.2102MCV: mean corpuscular volume; MCH: mean corpuscular hemoglobin; Hb: Hemoglobin. Variant annotations were performed using the following reference transcript sequences: HBA1: NM_000558.5, HBA2: NM_000517.6, and HBB: NM_000518.5. The phenotypes of 6 novel variants which predicted to be pathogenic. MCV: mean corpuscular volume; MCH: mean corpuscular hemoglobin; Hb: Hemoglobin. Variant annotations were performed using the following reference transcript sequences: HBA1: NM_000558.5, HBA2: NM_000517.6, and HBB: NM_000518.5. Discussion The global public health burden of thalassemia, a group of inherited blood disorders, is significant due to its high prevalence and diverse clinical manifestations (Tuo et al., 2024). Most thalassemia carriers can typically be identified through standard hematological analyses of mean corpuscular volume (MCV) and mean corpuscular hemoglobin (MCH), supplemented by traditional genetic testing. Methods for detecting thalassemia include restriction fragment length polymorphism (RFLP), polymerase chain reaction-reverse dot blotting (PCR-RDB), real-time quantitative PCR (Realtime-PCR), Sanger sequencing, and multiplex ligation-dependent probe amplification (MLPA) (Jiang et al., 2022; Zhong et al., 2023; Wang X. et al., 2023). However, these previous approaches mainly focus on known target variants, limiting their effectiveness in identifying rarer deletions, point mutations, or novel mutations in thalassemia-related genes. Moreover, due to the distinct mutational mechanics of the HBA1, HBA2, and HBB genes, a single conventional test is often insufficient for a comprehensive assessment of variations across these globin genes (Rehman et al., 2021; Alhuthali et al., 2023). Next-generation sequencing (NGS), which is increasingly applied for diagnosing various genetic disorders, offers advantages over traditional methods in detecting thalassemia. Yang et al. (Yang et al., 2023) provided valuable population-scale NGS data on thalassemia carrier frequencies in Southern China, which further supports the necessity of broad-spectrum genotyping approaches. Cao et al. (Cao et al., 2026) observed that NGS enhanced diagnostic yield in patients with low mean corpuscular volume compared to conventional testing, with the greatest improvement seen in cases of severe microcytic hypochromic anemia. Previous studies also suggest that NGS may serve as a first-line test for genetic screening of thalassemia (Achour et al., 2021), demonstrating improved performance compared to conventional screening techniques, a notion backed by a premarital screening data from the Chinese population (Zhang et al., 2019). In this study, we employed NGS for large-scale population screening to assess the frequency of rare thalassemia carriers in the Ganzhou region. This research marks a advancement in understanding the prevalence and variety of rare thalassemia variants in endemic areas, providing valuable insights for public health strategies. Our findings indicate that 1.29% of the screened population carried rare thalassemia variants, underscoring the necessity for targeted screening and prevention measures in regions with elevated carrier rates. Thalassemia is highly heterogeneous (Musallam et al., 2023). While some earlier studies have cautioned against the possibility of missed detections of rare thalassemia through prevention and control measures, they failed to offer specific details about the types, frequency, and distribution of rare thalassemia to enhance preventive efforts. Additionally, they did not indicate whether there are regional differences in this regard (Zhang et al., 2019; Zulkeflee et al., 2023). The Ganzhou area had a unique distribution pattern of rare thalassemia. Our findings indicated significant geographical differences in distribution. The areas of Nan Kang, Xingguo, and Zhanggong, situated to the northwest of Ganzhou, warrant increased focus in the subsequent phases of thalassemia prevention and control efforts. Given Ganzhou’s substantial population of approximately 8,989,200 and an area spanning 39,379.64 km2 (https://baike.so.com/doc/3631974-3817930.html), situated in the central region of southern China, investigating the genotype and distribution of rare thalassemia in this locale is critical for establishing a theoretical foundation for effective intervention strategies. Recent research has emphasized the impact of population migration on thalassemia prevention efforts. While the prevalence of thalassemia remains high in the southern regions, the northern areas have reported significantly fewer cases (Yang et al., 2019; Li, 2017). However, as a result of population movements, the number of thalassemia patients in northern China has been rising (He et al., 2021; Zhou et al., 2021). Therefore, identifying rare thalassemia cases also holds significant implications for broader prevention and control measures, including prenatal diagnosis and genetic counseling, particularly as our findings indicate that most individuals carrying these rare variants fall within the reproductive age group. This scenario also highlights an urgent need for targeted prevention strategies for rare thalassemia in Ganzhou. Notably, the top nine genotypes identified in rare α/β-thalassemia accounted for more than 80% of cases. Incorporating these genotypes into traditional thalassemia screening methods could represent a considerably improved strategy for detection. Some of the rare variants identified include hemoglobin variants, which have been shown to correlate with unstable hemoglobins that can lead to varying degrees of hemolytic anemia (Zhang J. et al., 2025). When combined with thalassemia, this could result in more severe manifestations of anemia (Du et al., 2021). The phenotypic expression of thalassemia is not solely determined by the presence or absence of a mutation but is modulated by multiple factors, including gene dosage, the specific nature of the variant (β0 vs. β+), co-inheritance of α- and β-thalassemia, and potentially other genetic modifiers. Furthermore, a significant genotype-phenotype discrepancy should serve as a clinical red flag, prompting further investigation. Such discrepancies may indicate the presence of additional undetected variants (e.g., rare CNVs or mutations in genes), alternative diagnoses (e.g., co-existing iron deficiency anemia or other hemoglobinopathies), or technical errors in genotyping. Additionally, we discovered 23 novel types of thalassemia variants that had not been previously reported in databases such as HGMD, HbVar, or PubMed. According to ACMG guidelines, most of these variants were categorized as variants of uncertain significance (VUS), suggesting they are more likely benign mutations predominantly located in the 5′UTR and 3′UTR regions. Among these, six variants were classified as “Likely Pathogenic,” prompting us to collect clinical hematological data for these cases. The HBB:c.50G>T (Gly > Val) mutation (PM2_P, PP3, PM3, PM5) could be classified as a severe variant based on its association with clinical symptoms. In contrast, the phenotype associated with the HBB:c.50G>A (Gly > Asp) variant, as described in existing literature, appears to be clinically silent despite the presence of an abnormal hemoglobin band accounting for approximately 40%–45% of the total hemoglobin. (Arribalzaga et al., 1996), a discrepancy likely linked to the differing polarities of the mutated amino acids: valine (hydrophobic) versus aspartic acid (polar). In addition, several limitations warrant consideration. Our CNV detection was limited to known deletions prevalent in the Chinese population, including -α3.7, -α4.2, --SEA, --THAI, FIL, Chinese-type Gγ+(Aγδβ)0, Taiwanese-type, and SEA-HPFH. While this panel covers the vast majority of clinically relevant CNVs in this region, we cannot completely exclude the possibility of other rare or atypical CNVs. Importantly, any discordance between genotype and clinical phenotype would prompt further investigation using dedicated CNV detection platforms such as long-read sequencing. In conclusion, we conducted a comprehensive study on rare and newly identified thalassemia variants in the Ganzhou area for the first time, leveraging a large sample size to explore the diversity and complexity of the condition. The enhanced diagnostic capabilities provided by NGS were essential to this research. Our findings contribute valuable data on rare thalassemia genotypes and serve as important references for future prevention and control efforts, both in this region and in other areas with high prevalence rates.