Abstract
Aims
To investigate the expression of acidic leucine-rich nuclear phosphoprotein 32 family member A (ANP32A) in extranodal NK/T-cell lymphoma (ENKTL) and to evaluate its relationship with clinicopathological characteristics, proliferative activity, and prognosis.
Methods
A total of 63 patients with ENKTL diagnosed at the First Affiliated Hospital of Zhengzhou University between 2012 and 2021 were retrospectively included. Reactive lymphoid tissue from the nasopharynx was used as the control tissue. Immunohistochemistry was performed to detect ANP32A expression in ENKTL and control tissues. The association between ANP32A expression and clinicopathological characteristics, Ki-67 index, and patient survival was analyzed. Kaplan–Meier survival analysis, univariate and multivariate Cox regression analyses, and receiver operating characteristic (ROC) curve analysis were performed to assess the prognostic significance of ANP32A.
Results
Among the 63 patients with ENKTL, there were 42 males and 21 females, with a median age of 50 years (range, 12–80 years). Compared with control tissues, ANP32A expression was increased in ENKTL tissues, and positive staining was mainly localized in the nuclei of tumor cells. According to the modified H-score, 23 cases showed high ANP32A expression and 40 cases showed low expression. High ANP32A expression was significantly associated with advanced Ann Arbor stage, higher Prognostic Index for Natural Killer Lymphoma (PINK) score, elevated β2-microglobulin level, and high Ki-67 index (all P < 0.05). Spearman correlation analysis showed a weak positive correlation between ANP32A expression and Ki-67 index (r = 0.330, P < 0.01). Kaplan–Meier analysis showed that patients with high ANP32A expression had significantly shorter overall survival (OS) than those with low expression (log-rank P < 0.0001). Univariate Cox regression analysis showed that high ANP32A expression was associated with poor prognosis (hazard ratio(HR) = 4.55; 95% confidence interval (CI), 2.50–8.52; P < 0.001). Multivariate Cox regression analysis further demonstrated that ANP32A high expression remained an independent adverse prognostic factor (HR = 3.61, 95% CI: 1.95–6.89, P < 0.001). ROC analysis showed that the area under the curve (AUC) of ANP32A alone for predicting OS was 0.776, which was higher than that of Ann Arbor stage alone (AUC = 0.667); when combined with Ann Arbor stage, the AUC increased to 0.800.
Conclusions
ANP32A is increased in ENKTL and is associated with aggressive clinicopathological features, increased proliferative activity, and poor survival outcome. ANP32A may serve as a useful supplementary prognostic marker in ENKTL and may provide additional prognostic information when combined with Ann Arbor stage.
Supplementary Information
The online version contains supplementary material available at https://doi.org/10.1186/s13000-026-01801-y.
Keywords: Extranodal NK/T-cell lymphoma, ANP32A, immunohistochemistry, prognosis, Ki-67
Introduction
Extranodal NK/T-cell lymphoma (ENKTL) is a mature NK-T-cell neoplasm closely associated with Epstein–Barr virus infection. It is highly aggressive, shows a distinct geographic distribution, and is more prevalent in Asia and Central and South America; the lesions most commonly arise in the nasal cavity and upper aerodigestive tract [1–3]. In recent years, survival has improved in a subset of patients with the introduction of asparaginase-based regimens, immunotherapy, and other novel therapeutic strategies. However, the outcomes of relapsed or refractory cases remain unsatisfactory, indicating that ENKTL is still a highly aggressive malignancy with rapid clinical progression and marked prognostic heterogeneity [4–7].
At present, Ann Arbor staging and the Prognostic Index for Natural Killer Lymphoma (PINK) score remain commonly used tools for prognostic evaluation in patients with ENKTL. Nevertheless, survival differences can still be observed among patients within the same risk group, suggesting that existing clinical parameters alone are insufficient to fully distinguish patients with different outcomes [8, 9]. By integrating genomic and transcriptomic data, Xiong et al. classified ENKTL into three molecular subtypes: tumor suppressor–immune modulator (TSIM), MGA-BRDT (MB), and HDAC9-EP300-ARID1A (HEA). These subtypes differed in EBV-related features, genetic alterations, immune characteristics, and clinical outcomes, and this classification was independent of conventional clinical scoring systems, including IPI, PINK, and PINK-E [5, 10]. The TSIM subtype is characterized by alterations involving tumor suppressor and immune-modulatory pathways, the MB subtype is associated with MYC-related biology and relatively unfavorable outcomes, and the HEA subtype is enriched for epigenetic alterations. This molecular classification provides important biological insight beyond conventional clinical staging and prognostic scoring systems. In addition, previous studies have shown that, beyond clinical scoring systems, some protein biomarkers associated with tumor biological behavior can also provide prognostic information; for example, high PLK1-p-PLK1 expression has been linked to poorer survival [11]. Other studies have summarized the progress in histological and molecular biomarkers in ENKTL and suggested that these markers may complement prognostic assessment beyond existing clinical scores [12, 13]. Therefore, identifying molecular biomarkers that can be detected in routine pathology practice and that supplement existing clinical indicators is of practical importance.
ANP32A (acidic leucine-rich nuclear phosphoprotein 32 family member A) is a member of the ANP32 family and is primarily localized in the nucleus. ANP32A was selected as an exploratory but biologically motivated marker in the present study. Previous studies have shown that ANP32A is involved in chromatin-associated regulation, histone H3 acetylation, transcriptional regulation, and tumor cell proliferation. In acute myeloid leukemia, ANP32A is upregulated and promotes leukemic cell proliferation and survival through regulation of histone H3 acetylation [14]. In acute megakaryoblastic leukemia and multiple myeloma, ANP32A has also been implicated in tumor cell growth and disease progression [15, 16]. These findings suggest that ANP32A may participate in tumor progression in several hematologic malignancies. Because ANP32A is primarily a nuclear protein, the expected immunohistochemical staining pattern is predominantly nuclear. In non-neoplastic lymphoid tissue, scattered lymphoid cells may show weak to moderate nuclear staining; therefore, in the present study, ANP32A scoring was restricted to tumor cells. However, the expression pattern and clinical significance of ANP32A in ENKTL remain unclear.
Based on these observations, the present study examined ANP32A expression in extranodal NK/T-cell lymphoma, analyzed its association with clinicopathological characteristics, Ki-67 index, and patient survival, and further evaluated its prognostic value when combined with Ann Arbor stage, in order to clarify the clinical significance of ANP32A in ENKTL and to provide a potentially useful histological indicator for prognostic assessment.
Materials and methods
Patient selection and clinicopathological data collection
This was a single-center retrospective study including 63 patients with ENKTL diagnosed at the First Affiliated Hospital of Zhengzhou University between 2012 and 2021. All cases were pathologically diagnosed according to the World Health Organization (WHO) classification criteria for hematopoietic and lymphoid tumors and were re-reviewed according to the updated 2020 criteria. The inclusion criteria were the availability of formalin-fixed paraffin-embedded (FFPE) tissue specimens for immunohistochemical analysis and sufficient clinicopathological and follow-up data for survival analysis, including at least age, sex, primary site, Ann Arbor stage, PINK score, survival status, and follow-up time. Laboratory parameters such as LDH, β2-microglobulin, and blood EBV-DNA were analyzed based on availability and were not mandatory for inclusion. Reactive lymphoid tissue from the nasopharynx was used as the control tissue.
Clinicopathological data collected from medical records included age, sex, primary site, B symptoms, Ann Arbor stage, PINK score, serum lactate dehydrogenase (LDH), serum β2-microglobulin (β2-MG), blood Epstein–Barr virus DNA (EBV-DNA), and Ki-67 index. According to the anatomical site of the primary lesion, cases were classified as nasal type or extranasal type. Elevated LDH was defined as > 245 U/L (reference range, 75–245 U/L), elevated β2-MG was defined as > 3 mg/L (reference range, 0–3 mg/L), and positive blood EBV-DNA was defined as > 5.00 × 10². The PINK score was calculated according to previously published criteria [9]. For statistical analysis, patients were further grouped into PINK scores of 0–2 and 3–4, and the Ki-67 index was categorized as ≤ 70% and > 70%.
This study was approved by the Ethics Committee of the First Affiliated Hospital of Zhengzhou University (Approval No. 2025-KY-1374).
Immunohistochemical staining
FFPE tissue specimens from 63 ENKTL cases were sectioned at 3 μm thickness. After routine deparaffinization, rehydration, antigen retrieval in Tris-EDTA buffer, and blocking of endogenous peroxidase activity, immunohistochemical staining was performed according to standard procedures. ANP32A was detected using a polyclonal antibody (Proteintech, Cat. No. 15810-1-AP, 1:300 dilution), and Ki-67 was detected using an antibody from ZSGB-BIO (Cat. No. 261202, 1:1500 dilution). Immunoreactivity was visualized with DAB, followed by hematoxylin counterstaining.
Evaluation of immunohistochemical staining
ANP32A immunoreactivity was predominantly localized in the nuclei of tumor cells. For scoring, only nuclear staining in tumor cells was evaluated. Weak non-nuclear staining or nonspecific background staining, if present, was not included in the modified H-score assessment. All slides were independently reviewed by two pathologists who were blinded to the clinical data. For each case, one representative FFPE block containing sufficient viable tumor tissue was selected for immunohistochemical analysis. In cases of discrepant scoring, a senior pathologist reviewed the slides and made the final decision.
ANP32A expression was evaluated using a modified H-score method. The proportion of positively stained tumor cells at each staining intensity was recorded as a value from 0 to 1, and staining intensity was scored as 0, 1, 2, or 3. The final score was calculated as the sum of the products of the staining intensity score and the corresponding proportion of positive cells, i.e., modified H-score = Σ(Pi × Ii), where Pi represents the proportion of positive tumor cells at each staining intensity and Ii represents the corresponding staining intensity. Therefore, the final score ranged from 0 to 3. Cases with a score > 2 were defined as having high ANP32A expression, whereas those with a score ≤ 2 were defined as having low ANP32A expression.
The Ki-67 index was defined as the percentage of tumor cell nuclei showing positive staining and was categorized as ≤ 70% or > 70% for statistical analysis.
Follow-up and survival analysis
Survival data were obtained from medical records, outpatient follow-up, and telephone follow-up. No structured interview or questionnaire was used in this study. The follow-up deadline was April 2025. Overall survival (OS) was defined as the interval from the date of diagnosis to death or the last follow-up. Patients who were alive at the last follow-up were considered censored.
Statistical analysis
GraphPad Prism 10.0 was used for survival analysis, Spearman correlation analysis, and Cox regression analysis, while SPSS 23.0 was used for chi-square tests, Fisher’s exact tests, and receiver operating characteristic (ROC) curve analysis. The association between ANP32A expression and clinicopathological parameters was analyzed using the chi-square test or Fisher’s exact test, as appropriate. The correlation between ANP32A staining score and Ki-67 index was assessed using Spearman’s rank correlation analysis.
Survival curves were generated using the Kaplan–Meier method and compared using the log-rank test. Univariate and multivariate Cox proportional hazards regression models were used to identify prognostic factors for OS in patients with ENKTL. Given the limited number of events and the potential overlap among some variables, not all significant variables identified in the univariate analysis were entered into the multivariate model simultaneously. Variables were selected for the multivariate Cox regression model based on univariate results, clinical relevance, and avoidance of excessive model complexity. ROC curve analysis was performed to compare the prognostic discrimination ability of ANP32A, Ann Arbor stage, and the combination of both. All tests were two-sided, and P < 0.05 was considered statistically significant.
Results
Clinicopathological characteristics of the patients
A total of 63 patients with extranodal NK/T-cell lymphoma (ENKTL) were retrospectively included in this study. The patients ranged in age from 12 to 80 years, with a median age of 50 years. Among them, 23 patients (36.5%) were older than 60 years, whereas 40 patients (63.5%) were 60 years of age or younger. There were 42 male patients (66.7%) and 21 female patients (33.3%).
The primary site was predominantly of the nasal type, accounting for 52 cases (82.5%). Among the 11 extranasal cases, the primary sites included the gastrointestinal tract in 6 cases, lung in 3 cases, and skin in 2 cases. The gastrointestinal tract cases involved the small intestine (n = 2), transverse colon (n = 1), ileocecal region (n = 2), and ileum (n = 1). B symptoms were present in 28 patients (44.4%). According to the Ann Arbor staging system, 37 patients (58.7%) were classified as stage I–II and 26 patients (41.3%) as stage III–IV. Based on the PINK score, 10 patients (15.9%) were classified into the high-risk group with scores of 3–4.
Some laboratory parameters were unavailable in a proportion of cases. Among patients with complete records, elevated LDH was observed in 24 of 53 cases (45.2%), elevated β2-microglobulin (β2-MG) in 15 of 52 cases (28.8%), and positive EBV-DNA in 9 of 46 cases (19.6%). The relatively low EBV-DNA positivity rate should be interpreted cautiously because EBV-DNA in this study referred specifically to peripheral blood EBV-DNA rather than tissue EBER status. In addition, because this was a retrospective cohort spanning 2012 to 2021, peripheral blood EBV-DNA testing was not uniformly performed at the same disease time point in all cases, and data were missing in 17 patients. Therefore, peripheral blood EBV-DNA was analyzed only in patients with available data. Molecular profiling data were not available for this retrospective cohort; therefore, correlation between ANP32A expression and the TSIM, MB, or HEA molecular subtypes could not be performed. PD-L1 expression was also not systematically assessed. Nineteen patients (30.2%) showed a Ki-67 index > 70%. Treatment information was available for all patients. All patients received asparaginase-based chemotherapy, and 3 patients also received radiotherapy. No patient received CHOP/CHOP-like regimen as the main treatment. The clinicopathological characteristics of the patients are summarized in Table 1.
Table 1.
Clinicopathological characteristics of 63 patients with ENKTL
| Characteristic | No. of patients (%) |
|---|---|
| Age(years) | |
| ≤60 | 40(63.5) |
| >60 | 23(36.5) |
| Sex | |
| Male | 42(66.7) |
| Female | 21(33.3) |
| B symptoms | |
| No | 35(55.6) |
| Yes | 28(44.4) |
| PINK score | |
| 0 | 13(20.6) |
| 1 | 22(34.9) |
| 2 | 18(28.6) |
| 3 | 8(12.7) |
| 4 | 2(3.2) |
| Ann Arbor stage | |
| I | 14(22.2) |
| II | 23(36.5) |
| III | 9(14.3) |
| IV | 17(27.0) |
| Primary site | |
| Nasal type | 52(82.5) |
| Extranasal type | 11(17.5) |
| Ki-67 index | |
| ≤ 70% | 44(69.8) |
| > 70% | 19(30.2) |
| LDH level | |
| Normal | 29(46.0) |
| Elevated | 24(38.1) |
| Missing | 10(15.9) |
| β2-MG level | |
| Normal | 37(58.7) |
| Elevated | 15(23.8) |
| Missing | 11(17.5) |
| Peripheral blood EBV-DNA status | |
| Negative | 37(58.7) |
| Positive | 9(14.3) |
| Missing | 17(27.0) |
Abbreviations: LDH Lactate dehydrogenase, β2-MG Beta-2 microglobulin, EBV-DNA Epstein–Barr virus DNA, PINK Prognostic Index for Natural Killer Lymphoma
ANP32A expression is increased in ENKTL and is associated with aggressive clinicopathological features
Immunohistochemical staining showed that ANP32A expression was higher in ENKTL tissues than in control tissues (Fig. 1A-B), and positive staining was predominantly localized in the nuclei of tumor cells. Across the cohort, ANP32A expression showed inter-case heterogeneity in both staining intensity and the proportion of positive tumor cells. ANP32A-high cases generally showed diffuse moderate to strong nuclear staining involving approximately 70%–95% of tumor cells, whereas ANP32A-low cases usually showed weak to moderate nuclear staining or focal strong nuclear staining involving approximately 20%–70% of tumor cells. Some cases showed heterogeneous staining within the tumor area, and this heterogeneity was incorporated into the modified H-score by recording the proportion of tumor cells at each staining intensity. According to the predefined cutoff described in the Methods, 23 cases (36.5%) showed high ANP32A expression, whereas 40 cases (63.5%) showed low expression.
Fig. 1.
ANP32A expression in ENKTL tissues. A Representative immunohistochemical staining of ANP32A in reactive lymphoid tissue and ENKTL tissues. ANP32A immunoreactivity was mainly localized in the nuclei of tumor cells. Scale bar = 200 μm. B ANP32A expression was higher in ENKTL tissues than in control tissues. P values are indicated in the figure. Abbreviations: ANP32A, acidic leucine-rich nuclear phosphoprotein 32 family member A; ENKTL, extranodal NK/T-cell lymphoma
Further analysis of the association between ANP32A expression and clinicopathological parameters (Table 2) showed that high ANP32A expression was associated with several aggressive clinicopathological features. In terms of stage distribution, high ANP32A expression was observed in 7 of 37 early-stage cases and 16 of 26 advanced-stage cases, whereas low ANP32A expression was observed in 30 of 37 early-stage cases and 10 of 26 advanced-stage cases. Regarding proliferative activity, high ANP32A expression was present in 11 of 44 cases with a Ki-67 index ≤ 70% and 12 of 19 cases with a Ki-67 index > 70%.
Table 2.
Correlation of high and low ANP32A expression with clinicopathological characteristics in patients with ENKTL
| Clinical parameter | Total n(%) | Expression of ANP32A | P value | |
|---|---|---|---|---|
| High(n = 23) | Low(n = 40) | |||
| Age(years) | 0.829 | |||
| ≤60 | 40(63.5) | 15 | 25 | |
| >60 | 23(36.5) | 8 | 15 | |
| Sex | ||||
| Male | 42(66.7) | 17 | 25 | 0.355 |
| Female | 21(33.3) | 6 | 15 | |
| B symptoms | ||||
| No | 35(55.6) | 9 | 26 | 0.056 |
| Yes | 28(44.4) | 14 | 14 | |
| PINK score | ||||
| 0–2 | 53(84.1) | 15 | 38 | 0.002 |
| 3–4 | 10(15.9) | 8 | 2 | |
| Ann Arbor stage | < 0.001 | |||
| I-II | 37(58.7) | 7 | 30 | |
| III-IV | 26(41.3) | 16 | 10 | |
| Primary site | ||||
| Nasal type | 52(82.5) | 16 | 36 | 0.081 |
| Extranasal type | 11(17.5) | 7 | 4 | |
| Ki-67 index | ||||
| ≤ 70% | 44(69.8) | 11 | 33 | 0.004 |
| > 70% | 19(30.2) | 12 | 7 | |
| LDH level* | ||||
| Normal | 29/53(54.7) | 6 | 23 | 0.051 |
| Elevated | 24/53(45.3) | 11 | 13 | |
| Missing | 10(15.9) | 6 | 4 | |
| β2-MG level* | 0.002 | |||
| Normal | 37/52(71.2) | 7 | 30 | |
| Elevated | 15/52(28.8) | 10 | 5 | |
| Missing | 11(17.5) | 6 | 5 | |
| Peripheral blood EBV-DNA status* | 0.241 | |||
| Negative | 37/46(80.4) | 11 | 26 | |
| Positive | 9/46(19.6) | 5 | 4 | |
| Missing | 17(27.0) | 7 | 10 | |
*P values for LDH, β2-MG, and EBV-DNA were calculated based on cases with available data; missing cases were not included in the statistical analysis.
Data in bold indicate statistically significant values (P < 0.05)
Notably, discordant cases were observed in the combined ANP32A expression and Ann Arbor stage analysis. Among the 7 ANP32A-high early-stage cases, 6 patients died during follow-up, and all six had OS shorter than the cohort median OS of 50.0 months. Conversely, among the 10 ANP32A-low advanced-stage cases, 4 patients died during follow-up; among these, 3 had OS shorter than the cohort median OS, and 1 had an OS of 50.0 months. These discordant cases indicate that ANP32A expression should not be interpreted as a standalone prognostic determinant, but rather as a supplementary marker that may refine, but cannot replace, established clinicopathological risk assessment.
The frequency of ANP32A high expression was higher in advanced-stage patients (stage III-IV) than in early-stage patients (stage I-II) (χ² = 14.76, P < 0.001). In addition, high ANP32A expression was associated with a higher PINK score (scores 3–4, χ² = 9.70, P = 0.002), a higher Ki-67 index (> 70%, χ² = 8.34, P = 0.004), and elevated β2-MG levels (Fisher’s exact test, P = 0.002). In contrast, no significant differences in ANP32A expression were observed according to sex, age, B symptoms, primary site, LDH level, or peripheral blood EBV-DNA status (all P > 0.05).
ANP32A expression was weakly correlated with the Ki-67 index in ENKTL
To investigate the association between ANP32A expression and proliferative activity in ENKTL, we further evaluated the relationship between ANP32A staining score and Ki-67 index. Immunohistochemical staining showed that cases with high ANP32A expression were generally accompanied by a higher Ki-67 index, whereas cases with low ANP32A expression tended to show a lower Ki-67 index (Fig. 2A).
Fig. 2.
Association between ANP32A expression and Ki-67 index in ENKTL. A Representative immunohistochemical staining showing ANP32A expression and Ki-67 index in ENKTL cases with different proliferative activity (Scale bar = 200 μm). B Spearman correlation analysis between ANP32A staining score and Ki-67 index. ANP32A expression showed a weak positive correlation with the Ki-67 index. C Comparison of ANP32A expression scores between cases with low Ki-67 index (≤ 70%) and high Ki-67 index (> 70%). ANP32A expression scores were higher in the high Ki-67 group. The P value is indicated in the figure. Abbreviations: ANP32A, acidic leucine-rich nuclear phosphoprotein 32 family member A; ENKTL, extranodal NK/T-cell lymphoma
Spearman correlation analysis demonstrated a weak positive correlation between the ANP32A staining score and the Ki-67 index (r = 0.330, P < 0.01) (Fig. 2B). In addition, ANP32A expression scores were higher in the high Ki-67 group (> 70%) than in the low Ki-67 group (≤ 70%) (Fig. 2C). These findings suggest a modest association between ANP32A expression and proliferative activity in ENKTL.
High ANP32A expression was associated with poorer survival outcomes
Kaplan–Meier survival analysis showed that patients with high ANP32A expression had shorter OS than those with low ANP32A expression (log-rank P < 0.0001, Fig. 3A). The study endpoint was OS. During follow-up, 29 deaths occurred and 34 patients were censored. The median OS for the whole cohort was 50.0 months (95% CI, 19.0 months to not estimable). Analysis according to Ann Arbor stage showed that patients with advanced-stage disease (stage III–IV) had shorter OS than those with early-stage disease (stage I–II) (log-rank P = 0.002, Fig. 3B). Univariate Cox regression analysis showed that high ANP32A expression was associated with poor prognosis (HR = 4.55, 95% CI: 2.50–8.52, P < 0.001) (Table 3). In addition, B symptoms, PINK score, Ann Arbor stage, primary site, Ki-67 index, LDH level, and β2-MG level were also associated with OS (all P < 0.05, Table 3).
Fig. 3.
Survival analysis according to ANP32A expression and Ann Arbor stage in patients with ENKTL. A Kaplan–Meier OS curves of patients with high and low ANP32A expression. B Kaplan–Meier OS curves of patients with early-stage (I–II) and advanced-stage (III–IV) disease according to Ann Arbor stage. C Kaplan–Meier OS curves of patients stratified by the combined ANP32A expression and Ann Arbor stage groups. The overall log-rank P value is shown in the figure. Abbreviations: ANP32A, acidic leucine-rich nuclear phosphoprotein 32 family member A; ENKTL, extranodal NK/T-cell lymphoma; OS, overall survival
Table 3.
Univariate and multivariate Cox regression analyses of OS in patients with ENKTL
| Variable | Univariate analysis | Multivariate analysis | |||
|---|---|---|---|---|---|
| HR (95% CI) | P value | HR (95% CI) | P value | ||
| ANP32A (High vs. Low) | 4.55 (2.50–8.52) | < 0.001 | 3.61(1.95–6.89) | < 0.0001 | |
| Age (> 60 vs. ≤60 years) | 0.94(0.40–2.01) | 0.87 | - | - | |
| Sex (Male vs. Female) | 0.73(0.35–1.61) | 0.43 | - | - | |
| B symptoms (Yes vs. No) | 2.81(1.33–6.33) | 0.006 | 2.29(1.06–5.22) | 0.033 | |
| PINK score(3–4 vs. 0–2) | 4.87(2.01–11.17) | < 0.001 | - | - | |
| Ann Arbor stage (III–IV vs. I–II) | 2.95(1.41–6.37) | 0.004 | 2.20(0.99-5.00) | 0.052 | |
| Primary site( Extranasal type vs. Nasal type ) | 3.33(1.36–7.43) | 0.01 | - | - | |
| Ki-67 index(> 70% vs. ≤70%) | 2.93(1.39–6.13) | 0.005 | - | - | |
| LDH level (Elevated vs. Normal) | 3.50(1.47–9.19) | 0.004 | - | - | |
| β2-MG level (Elevated vs. Normal) | 3.79(1.56–9.01) | 0.004 | - | - | |
| Peripheral blood EBV-DNA status (Positive vs. Negative) | 1.84(0.65–4.54) | 0.23 | - | - | |
Data in bold indicate statistically significant values (P < 0.05)
To further evaluate the relationship between ANP32A expression and Ann Arbor stage, stratified survival analysis was performed. Within the same stage group, patients with low ANP32A expression tended to show better survival than those with high ANP32A expression. Kaplan–Meier analysis of the combined ANP32A expression and Ann Arbor stage groups showed distinct OS patterns among the subgroups (Fig. 3C). These findings suggest that ANP32A may provide additional prognostic information within stage-defined subgroups, although this observation should be interpreted cautiously given the limited sample size.
ANP32A was an independent prognostic factor in ENKTL and may improve stage-based prognostic discrimination
To further evaluate the independent prognostic value of ANP32A, ANP32A expression, Ann Arbor stage, and B symptoms were included in the multivariate Cox regression model based on the univariate results and their clinical relevance. The results showed that high ANP32A expression (HR = 3.61, 95% CI: 1.95–6.89, P < 0.001) and the presence of B symptoms (HR = 2.29, 95% CI: 1.06–5.22, P = 0.033) were independent factors affecting OS in patients with ENKTL, whereas Ann Arbor stage showed borderline significance in the multivariate analysis (HR = 2.20, 95% CI: 0.99–5.00, P = 0.052) (Fig. 4A; Table 3).
Fig. 4.
Multivariate analysis and prognostic performance of ANP32A in ENKTL. A Forest plot of multivariate Cox proportional hazards regression analysis for OS in patients with ENKTL. B Receiver operating characteristic (ROC) curves for ANP32A, Ann Arbor stage, and the combination of both in predicting OS. The combination of ANP32A and Ann Arbor stage may provide improved prognostic discrimination compared with Ann Arbor stage alone. Abbreviations: ANP32A, acidic leucine-rich nuclear phosphoprotein 32 family member A; ENKTL, extranodal NK/T-cell lymphoma; OS, overall survival; ROC, receiver operating characteristic; AUC, area under the curve; HR, hazard ratio; CI, confidence interval
Subsequently, ROC curve analysis was performed to compare the discriminatory ability of different indicators for OS (Fig. 4B). The results showed that the area under the curve (AUC) of ANP32A alone for predicting OS was 0.776, which was higher than that of Ann Arbor stage alone (AUC = 0.667). When ANP32A was combined with Ann Arbor stage, the AUC further increased to 0.800. These findings suggest that the combination of ANP32A and Ann Arbor stage may provide additional prognostic information beyond Ann Arbor stage alone. A supplementary multivariate Cox regression analysis including ANP32A expression and PINK score was also performed, and the result is shown in Supplementary Table S1.
Discussion
In the present study, ANP32A was increased in ENKTL tissues, and high ANP32A expression was significantly associated with advanced Ann Arbor stage, higher PINK score, elevated β2-MG level, and high Ki-67 expression. Survival analysis further showed that patients with high ANP32A expression had significantly shorter OS. In multivariate Cox regression analysis, ANP32A remained associated with poor prognosis. These findings indicate that high ANP32A expression is linked to a heavier clinicopathological burden and an unfavorable survival outcome in ENKTL.
Ann Arbor stage and the PINK score remain commonly used clinical tools for prognostic assessment in patients with ENKTL. However, in routine practice, patients within the same stage or the same risk group may still experience different clinical outcomes [8, 9]. In the present study, the primary analysis focused on whether ANP32A could provide additional prognostic information beyond conventional stage-based stratification; therefore, Ann Arbor stage was retained as the principal staging comparator because it directly reflects disease extent. At the same time, PINK remains a clinically meaningful composite prognostic model in ENKTL [8, 9]. In a supplementary sensitivity analysis including PINK score in a multivariate Cox model, ANP32A remained significantly associated with OS, although this result should be interpreted cautiously given the limited sample size. These findings suggest that incorporation of ANP32A into existing clinical parameters may further refine prognostic discrimination. Previous studies have also shown that adding new variables to existing prognostic models is feasible [17]. For example, Kang et al. incorporated β2-microglobulin into PINK and established a modified prognostic model, further supporting the value of introducing additional indicators into conventional prognostic systems [8].
We also found that ANP32A expression was weakly positively correlated with the Ki-67 index. Ki-67 is a commonly used indicator of tumor cell proliferative activity, and previous studies have shown its prognostic value in ENKTL [18, 19]. Accordingly, in the present cohort, high ANP32A expression was more frequently observed in cases with a higher Ki-67 index, suggesting that ANP32A upregulation may be linked to active tumor cell proliferation. In addition to Ki-67, previous studies have reported that PLK1/p-PLK1, EZH2/H3K27me3, EGR1, and KMT2D alterations detected in ctDNA are associated with the outcome of patients with ENKTL [11, 20–25]. Together, these observations indicate that, beyond clinical scoring systems, histological and molecular markers may provide additional information for prognostic assessment [26, 27].
Evidence from other hematologic malignancies also supports the findings of the present study. Yang et al. reported that ANP32A is highly expressed in acute myeloid leukemia and regulates histone H3 acetylation, thereby promoting leukemic cell proliferation and survival [14]. Huang et al. further showed that high ANP32A expression was associated with adverse outcomes in AML [28]. In acute megakaryoblastic leukemia and multiple myeloma, ANP32A has likewise been implicated in tumor cell growth and disease progression [15, 16]. These findings suggest that ANP32A may function as a chromatin-associated regulator that supports malignant transcriptional programs in hematologic neoplasms. Although the precise molecular mechanism of ANP32A in ENKTL remains unclear, the positive association between ANP32A expression and Ki-67 index, together with its correlation with advanced stage and poorer survival, raises the possibility that ANP32A may contribute to ENKTL progression by promoting proliferation-related transcriptional activity and epigenetic dysregulation. Further functional studies are needed to clarify whether ANP32A acts through histone acetylation-related pathways or other nuclear regulatory mechanisms in ENKTL. It should be noted that, in the multivariate model constructed in this study, ANP32A expression and B symptoms remained statistically significant, whereas Ann Arbor stage showed only borderline significance. These findings suggest that incorporation of ANP32A into existing clinical parameters may further refine prognostic discrimination. In the present study, the combined ANP32A expression and Ann Arbor stage analysis was intended as an exploratory stratified survival analysis, and the observed subgroup differences should be interpreted cautiously given the limited sample size.
From a clinical perspective, ANP32A has the advantage of being readily assessable by immunohistochemistry. Compared with sequencing- or ctDNA-based approaches, immunohistochemistry can be more easily incorporated into routine pathological practice [12]. Although the present findings suggest that ANP32A may serve as a supplementary histological marker for prognostic evaluation in ENKTL, this conclusion should be interpreted cautiously and requires validation in larger, independent, and preferably multicenter cohorts before clinical application.
This study has several limitations. First, it was a single-center retrospective study with a limited sample size, and the conclusions still require validation in independent cohorts. In addition, ANP32A expression was evaluated mainly by immunohistochemistry, and orthogonal validation using an independent method was not performed. Some potentially relevant clinical variables, such as ECOG performance status, were also not consistently documented or reliably retrievable in this retrospective cohort, particularly in earlier cases, and therefore could not be included in the current analysis without substantially reducing the analyzable sample size. Another important limitation is the lack of molecular profiling data. Therefore, we were unable to evaluate the relationship between ANP32A expression and recently proposed molecular subtypes of ENKTL, including TSIM, MB, and HEA. In addition, PD-L1 expression was not systematically assessed, and peripheral blood EBV-DNA data were incomplete and not uniformly obtained at the same clinical time point. Future studies integrating immunohistochemistry, molecular classification, PD-L1 status, and EBV-DNA dynamics are needed to better define the biological and prognostic significance of ANP32A in ENKTL. Second, although the present study demonstrated clinicopathological and prognostic associations, it was mainly based on immunohistochemical findings and clinical follow-up data, and functional experiments were not performed to directly clarify the role of ANP32A in ENKTL. Nevertheless, our findings demonstrate a stable association between ANP32A expression and clinical progression and survival in ENKTL.
In conclusion, ANP32A is increased in ENKTL and is associated with high proliferative activity, aggressive clinicopathological features, and poor survival outcome. ANP32A may complement Ann Arbor stage–based prognostic stratification and has potential value as a supplementary prognostic marker in ENKTL.
Supplementary Information
Abbreviations
- ENKTL
Extranodal NK/T-cell lymphoma
- ANP32A
Acidic leucine-rich nuclear phosphoprotein 32 family member A
- WHO
World Health Organization
- FFPE
Formalin-fixed paraffin-embedded
- LDH
Lactate dehydrogenase
- β2-MG
Beta-2 microglobulin
- EBV
Epstein–Barr virus
- EBV-DNA
Epstein–Barr virus DNA
- PINK
Prognostic Index for Natural Killer Lymphoma
- OS
Overall survival
- ROC
Receiver operating characteristic
- AUC
Area under the curve
- HR
Hazard ratio
- CI
Confidence interval
- AML
Acute myeloid leukemia
- ctDNA
Circulating tumor DNA
Authors’ contributions
Yajun Huo performed most of the study, including case collection, follow-up, experiments, data analysis, figure preparation, and manuscript drafting. Zhiqi Zhang participated in part of the experiments and clinical data collection. Enjie Liu, Guannan Wang, and Wugan Zhao contributed to pathological data review and result verification. Dandan Zhang and Yanping Zhang contributed to clinical data checking and follow-up data completion. Yifan Shang, Huimin Du, and Jun Yu contributed to literature review, figure checking, and manuscript preparation. Chongli Zhang and Susu Lu provided technical support and specimen management. Wencai Li conceived and supervised the study, revised the manuscript critically, approved the final version, and takes responsibility for the integrity of the work. All authors read and approved the final manuscript.
Funding
This work was supported by the National Natural Science Foundation of China (Grant No. 82170184).
Data availability
The data that support the findings of this study are available from the corresponding author upon reasonable request.
Declarations
Ethics approval and consent to participate
The study was conducted in accordance with the Declaration of Helsinki and approved by the Ethics Committee of the First Affiliated Hospital of Zhengzhou University (Approval No. 2025-KY-1374). The requirement for informed consent was waived by the Ethics Committee because of the retrospective nature of the study and the use of archived tissue specimens and anonymized clinical data.
Competing interests
The authors declare no competing interests.
Footnotes
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
References
- 1.He X, Gao Y, Li Z, et al. Review on natural killer/T-cell lymphoma. Hematol Oncol. 2023;41(2):221–9. [DOI] [PubMed] [Google Scholar]
- 2.Tse E, Fox CP, Glover A, et al. Extranodal natural killer/T-cell lymphoma: An overview on pathology and clinical management. Semin Hematol. 2022;59(4):198–209. [DOI] [PubMed] [Google Scholar]
- 3.Chan WL, Hue SS, Deng L, et al. Extranodal NK/T-cell lymphoma: an update of the molecular characterization of the tumor and microenvironment, and its clinical implications. Lancet Reg Health West Pac. 2025;62:101550. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Tian XP, Cao Y, Cai J, et al. Novel target and treatment agents for natural killer/T-cell lymphoma . J Hematol Oncol. 2023;16(1):78. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Xiong J, Cui BW, Wang N, et al. Genomic and Transcriptomic Characterization of Natural Killer T Cell Lymphoma . Cancer Cell. 2020;37(3):403–e419406. [DOI] [PubMed] [Google Scholar]
- 6.Yan Z, Yao S, Wang Z, et al. Treatment of extranodal NK/T-cell lymphoma: From past to future . Front Immunol. 2023;14:1088685. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Fox CP, Civallero M, Ko YH, et al. Survival outcomes of patients with extranodal natural-killer T-cell lymphoma: a prospective cohort study from the international T-cell Project. Lancet Haematol. 2020;7(4):e284–94. [DOI] [PubMed] [Google Scholar]
- 8.Kang S, Cho H, Kim S, et al. A New Prognostic Index for Extranodal Natural Killer/T-Cell Lymphoma:Incorporation of Serum β-2 Microglobulin to PINK. Cancer Res Treat. 2023;55(1):314–24. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Kim SJ, Yoon DH, Jaccard A, et al. A prognostic index for natural killer cell lymphoma after non-anthracycline-based treatment: a multicentre, retrospective analysis. Lancet Oncol. 2016;17(3):389–400. [DOI] [PubMed] [Google Scholar]
- 10.Kim H, Ko YH. The Pathologic and Genetic Characteristics of Extranodal NK/T-Cell Lymphoma. Life (Basel), 2022,12(1). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Zhang Z, Liu E, Zhang D, et al. The expression and clinical significance of PLK1/p-PLK1 protein in NK/T cell Lymphoma. Diagn Pathol. 2023;18(1):129. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Chan JY, Lim JQ, Ong CK. Towards Next Generation Biomarkers in Natural Killer/T-Cell Lymphoma. Life (Basel), 2021,11(8). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Sun KM, Wong YH, Cheung KC et al. Update on Molecular Diagnosis in Extranodal NK/T-Cell Lymphoma and Its Role in the Era of Personalized Medicine . Diagnostics (Basel), 2022,12(2). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Yang X, Lu B, Sun X, et al. ANP32A regulates histone H3 acetylation and promotes leukemogenesis . Leukemia. 2018;32(7):1587–97. [DOI] [PubMed] [Google Scholar]
- 15.Sun X, Lu B, Han C, et al. ANP32A dysregulation contributes to abnormal megakaryopoiesis in acute megakaryoblastic leukemia . Blood Cancer J. 2017;7(12):661. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Wei R, Zhu Y, Zhang Y, et al. AIMP1 promotes multiple myeloma malignancy through interacting with ANP32A to mediate histone H3 acetylation . Cancer Commun (Lond). 2022;42(11):1185–206. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Huang D, Li Q, Li X, et al. The Novel Prognostic Index Model of Combining Circulating Tumor DNA and PINK-E Predicts the Clinical Outcomes for Newly Diagnosed Extranodal NK/T-cell Lymphoma . Hemasphere. 2023;7(1):e822. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Kim SJ, Kim BS, Choi CW, et al. Ki-67 expression is predictive of prognosis in patients with stage I/II extranodal NK/T-cell lymphoma, nasal type . Ann Oncol. 2007;18(8):1382–7. [DOI] [PubMed] [Google Scholar]
- 19.Jiang L, Li P, Wang H, et al. Prognostic significance of Ki-67 antigen expression in extranodal natural killer/T-cell lymphoma, nasal type . Med Oncol. 2014;31(10):218. [DOI] [PubMed] [Google Scholar]
- 20.Liu J, Liang L, Huang S, et al. Aberrant differential expression of EZH2 and H3K27me3 in extranodal NK/T-cell lymphoma, nasal type, is associated with disease progression and prognosis . Hum Pathol. 2019;83:166–76. [DOI] [PubMed] [Google Scholar]
- 21.Lee JY, Kim JH, Bang H, et al. EGR1 as a potential marker of prognosis in extranodal NK/T-cell lymphoma . Sci Rep. 2021;11(1):10342. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Li Q, Zhang W, Li J, et al. Plasma circulating tumor DNA assessment reveals KMT2D as a potential poor prognostic factor in extranodal NK/T-cell lymphoma . Biomark Res. 2020;8:27. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Kim WY, Jung HY, Nam SJ, et al. Expression of programmed cell death ligand 1 (PD-L1) in advanced stage EBV-associated extranodal NK/T cell lymphoma is associated with better prognosis . Virchows Arch. 2016;469(5):581–90. [DOI] [PubMed] [Google Scholar]
- 24.Feng Y, Feng X, Jing C, et al. The expression and clinical significance of programmed cell death receptor 1 and its ligand in tumor tissues of patients with extranodal nasal NK/T cell lymphoma . Sci Rep. 2022;12(1):36. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Bi C, Huang Y, Ali R, et al. MYC overexpression in natural killer cell lymphoma: prognostic and therapeutic implications . Haematologica. 2024;109(9):2810–21. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Tian XP, Zhang YC, Lin NJ, et al. Diagnostic performance and prognostic value of circulating tumor DNA methylation marker in extranodal natural killer/T cell lymphoma . Cell Rep Med. 2023;4(2):100859. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Kim JJ, Kim HY, Choi Z, et al. In-depth circulating tumor DNA sequencing for prognostication and monitoring in natural killer/T-cell lymphomas . Front Oncol. 2023;13:1109715. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Huang S, Huang Z, Ma C, et al. Acidic leucine-rich nuclear phosphoprotein-32A expression contributes to adverse outcome in acute myeloid leukemia . Ann Transl Med. 2020;8(6):345. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
The data that support the findings of this study are available from the corresponding author upon reasonable request.