Real-World Survival Outcomes of Atezolizumab and Bevacizumab in Unresectable Hepatocellular Carcinoma Patients With Hepatitis B Virus Infection: A Hospital-Based Cohort Study

Abstract

Background

Atezolizumab-Bevacizumab is one of the standard treatment for unresectable hepatocellular carcinoma patients. However, clinical data on the effectiveness of this regimen specifically in patients with HBV–related unresectable hepatocellular carcinoma remain limited.

Objectives

To evaluate survival outcomes and identify prognostic factors in HBV-infected unresectable HCC patients treated with Atezolizumab-Bevacizumab as first line systemic therapy.

Methods

A prospective combining retrospective cohort study, conducted on unresectable HCC patients with HBV infection, treated with Atezolizumab-Bevacizumab as the first-line systemic therapy at 3 major cancer-specialized hospitals in Northern Vietnam from June 2020 to August 2024.

Results

Totally 97 patients were included, the median age was 57 years, 89.4% were male. After a median follow-up of 11.1 (95% CI: 9.8-13.2) months, the median progression-free survival (mPFS) was 4.3 (95% CI: 3.4-5.2) months, and the median overall survival (mOS) was 20 (95% CI: 11.1-28.9) months. Multivariate analysis showed that Child-Pugh score A 6 points and B 7-8 points (HR: 2.0, P = 0.009 and HR: 4.5, P = 0.029), liver tumor size ≥80 mm (HR: 1.7, P = 0.037), and distant metastasis (HR: 1.9, P = 0.019), were independent predictors of PFS. Independent predictors of OS included ECOG - PS 1 (HR: 1.8, P = 0.047), ALBI grade 2-3 (HR: 2.0, P = 0.033), and NLR >2.9 (HR: 2.0, P = 0.023).

Conclusion

The Atezolizumab-Bevacizumab regimen has shown favorable effectiveness as first-line treatment for unresectable HBV-related HCC in Vietnam. These findings provide real-world evidence to support policymakers in developing treatment strategies in HBV-endemic regions and guide future research on optimizing HCC management.

Introduction

Hepatocellular carcinoma (HCC) is the most common primary liver cancer, accounting for 75-85% of liver cancer cases worldwide and ranking third among cancer deaths. According to Globocan 2022, 905.700 new cases and 830.200 deaths from HCC were recorded globally. ¹ It is expected that by 2040, the number of new cases and deaths will increase by more than 55%, reaching about 1.4 million cases and 1.3 million deaths annually. ² This disease burden poses a major challenge to the health system, especially in regions with high prevalence of chronic liver disease.

The risk factors for HCC include hepatitis B virus (HBV) and hepatitis C virus (HCV) infection, cirrhosis, alcoholism, obesity, metabolic syndrome, and non-alcoholic fatty liver disease (NAFLD).^3-6 Among of them, HBV and HCV are the major HCC risk factors with over 50% of HCC patients worldwide having HBV infection and 25% having HCV infection.^7,8 The prevalence of HBV and HCV varies by geographical region.^9,10 Vietnam has high incidence of HBV infection, 8-25% approximately, and HBV is also the main cause of HCC in Vietnam, with the rate of HCC patients with HBV reaching 65%.^11,12

Systemic therapy for unresectable HCC has advanced notably with the introduction of immunotherapy-based regimens. The IMbrave150 trial demonstrated that the combination of Atezolizumab, a PD-L1 immune checkpoint inhibitor, and Bevacizumab, an anti-VEGF monoclonal antibody, significantly improved overall survival compared to sorafenib (19.2 months vs 13.4 months). ¹³ These results led to the approval of this regimen as the first-line treatment for unresectable HCC in 2020. However, the IMbrave150 trial included a heterogeneous patient population, with only 49% of participants having HBV-related HCC, while the remaining 51% had HCV-related or non-viral etiologies and data on the real-world effectiveness of this regimen in patients with HBV-related HCC, especially in Vietnam or Asian regions, remain limited.

Given the high prevalence of HBV in Vietnam and the variability in patient responses to the Atezolizumab-Bevacizumab combination, it is crucial to investigate survival outcomes and identify prognostic factors specific to this subgroup. Moreover, understanding which clinical characteristics influence survival may help guide treatment decisions and improve patient outcomes. Therefore, this study aimed to evaluate the survival outcomes and determine predictive factors in Vietnamese patients with unresectable HBV-related HCC treated with the Atezolizumab-Bevacizumab regimen. The findings are expected to contribute real-world evidence and support more personalized treatment approaches in high-HBV-burden settings.

Materials and Methods

Design and Patient Characteristics

This hospital-based cohort study included both retrospective and prospective data. We conducted this study at 3 hospitals in Vietnam, on HBV-infected HCC patients treated with the Atezolizumab - Bevacizumab as the first-line systemic therapy from June 2020 to August 2024. Our study considered all patients who satisfied the following criteria: being diagnosed with HCC based on the criteria of the Vietnamese Ministry of Health’s guidelines; having hepatitis B virus infection (test HBV surface antigen (HBs-Ag) positive); ECOG performance status from 0-2 points; Child-Pugh A or Child-Pugh B liver function; having at least 1 target lesion on CT or MRI according to the Response Evaluation Criteria in Solid Tumors (RECIST). Patients were excluded if having at least one of the following criteria: co-infection with both HBV and HCV; autoimmune diseases or immunodeficiency; having a history of organ transplantation and taking anti-rejection drugs; symptomatic or uncontrolled hypertension or hypertensive encephalopathy; having history of bleeding due to esophageal varices within 6 months or untreated esophageal varices or gastric bleeding.

Pre-treatment data were collected include sex, age at treatment initiation, hepatitis status, history of previous local therapy, disease stage according to Barcelona classification, ECOG performance status, tumor characteristics and metastasis status (based on CT scan with contrast, MRI with contrast performed within 1 month before treatment), blood tests performed within 1 week before treatment, liver function classified according to Child-Pugh score, ALBI grade and NLR (the neutrophil-to-lymphocyte ratio). NLR was calculated as the absolute neutrophil count divided by the absolute lymphocyte count.

The study protocol was approved by the Ethics Committee of Hanoi Medical University (Decision No. 1072/GCN-HMUIRB, dated February 27, 2024) and by the leaders of all studied sites (3 hospitals). All prospective participants provided written informed consent prior to enrollment. For retrospective cases, informed consent was obtained through telephone contact with the patients or their family members (if patients died). Additionally, written approval for data usage was granted by the leaders of 3 hospitals. All patient information used in this study was fully de-identified prior to analysis in accordance with ethical standards, and no identifiable personal data were collected, stored, or reported.

The reporting of this study conforms to STROBE guidelines. ¹⁴

Treatment Characteristics and Assessment Methods

Patients received intravenous infusion of Atezolizumab 1200 mg and Bevacizumab 15 mg/kg body weight on the day 1 and then every 3 weeks. The regimen continued until disease progressed, no longer had clinical benefit or showed unacceptabled toxicity. Meanwhile these patients were given anti-virus B daily, supportive care as the major procedures. Patients were then switched to best supportive care, second-line systemic treatment, or local therapy depending on disease progression, liver function, and performance status if they had no response.

For outcome evaluation, the primary end points were PFS and OS. Treatment response was assessed according to RECIST 1.1 criteria every 3 months or as soon as the disease showed signs of progression. End of follow-up time was defined as the time of patient death or the last time the patient information was collected, or the cut-off point of this study (March 31, 2025).

Statistical Analysis

Descriptive statistics are presented as mean ± standard deviation (min-max), median (range, interquartile range [IQR]), or frequency (%). OS was calculated from the start of atezolizumab plus bevacizumab treatment until death from any cause. PFS was defined as the time from the start of treatment to progression or death from any cause. Survival outcomes were generated using the Kaplan–Meier method. Factors influencing OS and PFS were assessed using Cox proportional hazards regression models in which multivariate analysis used a stepwise backward elimination method (POUT = 0.10). Hazard ratios (HRs) and 95% confidence intervals (CIs) were reported. All P-values were two-sided, and P < 0.05 was considered statistically significant. Statistical analyses were performed using SPSS software (version 26.0; IBM Corp).

Results

Characteristics of Patients

According to Figure 1, from June 2020 to August 31, 2024, 176 HCC patients treated with Atezolizumab-Bevacizumab regimen were considered. After applying the inclusion and exclusion criteria, 97 patients were eligible and enrolled in the study.

Figure 1.

STROBE flow chart of patient selection. Abbreviations: HCC, Hepatocellular carcinoma; HBV, Hepatitis B virus; BCLC, Barcelona clinic liver cancer

According to the Table 1 above, the mean age was 57 (50 – 64); 90.7% of patients were male; 57.7% of patients had PS 0. Regarding liver function according to the Child-Pugh classification, most patients had Child-Pugh A (A 5 points: 56.7%; A 6 points: 40.2%), according to the ALBI grade 1, accounting for 43.3%. Patients with AFP ≥400 ng/ml accounted for 45.4%. At the pre-treatment, most patients were diagnosed at BCLC stage C (55.7%) with liver tumors >80 mm accounting for 61.5%; 48.5% of patients had portal vein thrombosis; 37.1% of studied patients had distant metastases.

Table 1.

Patient Demograpic (n = 97)

Variables	Number of patients	%
Sex
Male	88	90.7
Female	9	9.3
Age, median (IQR), years	57 (50 – 64)
ECOG
PS 0	56	57.7
PS 1	41	42.3
Pre-treatment staging
Recurrent, metastasis after radical treatment	16	16.5
BCLC B stage	12	12.4
BCLC C stage	54	55.7
Failure after TACE	15	15.4
Classification of child-Pugh score
A 5 score	55	56.7
A 6 score	39	40.2
B 7-8 score	3	3.1
ALBI grade
Grade 1	42	43.3
Grade 2-3	55	56.7
Classification of AFP level
<400 ng/ml	53	54.6
≥400 ng/ml	44	45.4
Liver tumor size (n = 91)
<80 mm	35	38.5
≥80 mm	56	61.5
Portal vein thrombosis
No	50	51.5
Yes	47	48.5
Distant metastases
No	61	62.9
Yes	36	37.1
Neutrophil (G/l)	5.0 ± 3.5	2.1 - 30.5
Lymphocytes (G/l)	1.9 ± 0.6	0.9 – 3.8

Abbreviations: ECOG, Eastern Cooperative Oncology Group; PS, performance status; BCLC, Barcelona Clinic Liver Cancer; TACE, Transarterial Chemoembolization; ALBI, albumin-bilirubin; AFP, alpha-fetoprotein; INR, International Normalized Ratio; IQR, interquartile range.

Response Rate

As shown in the Table 2, during the treatment course, 3 patients had complete response (3.1%), 21.6% of patients had partial response, 40.2% of patients had stable disease. Overall response rate: 24.7%; Disease control rate: 64.9%. The rate of progressive disease was 35.1%. DCR was 64.9% and ORR was 24.7%.

Table 2.

Clinical Response According to RECIST 1.1 (n = 97)

Treatment response	N	%
Complete response (CR)	3	3.1
Partial response (PR)	21	21.6
Stable disease (SD)	39	40.2
Progression disease (PD)	34	35.1

Progression Free Survival (PFS) and Overall Survival (OS)

In Figure 2 shows that the median progression-free survival (PFS) was 4.3 months (95% CI: 3.4-5.2). The progression-free survival rates at 3 months, 6 months, and 1 year were 65.3%; 37.1%, and 25.3%, respectively.

Figure 2.

Progression free survival (n = 97)

At the cut-off point of follow - up, 47/97 patients (48.5%) had died. As shown in Figure 3, the median overall survival was 20.0 months (95% CI: 11.1 – 28.9). The OS at 6-month, 1-year, and 2-year were 79.4%; 67.1%, and 45.1%, respectively.

Figure 3.

Overall survival (n = 97)

Predictive Factors Associated With Progression-Free Survival (PFS)

According to the Table 3 above, univariate Cox regression analysis in showed that the following factors: ECOG performance status, Child-Pugh score, ALBI grade, neutrophil/lymphocyte ratio (NLR), and liver tumor size were associated with PFS. Multivariate analysis showed that Child-Pugh score A-6 points and B 7-8 points (vs Child-Pugh score A-5 points; HR: 2.0, 95% CI: 1.2 - 3.4, P = 0.009 and HR: 4.5, 95% CI: 1.2 - 17.1, P = 0.029 respectively), liver tumor size ≥80 mm (vs < 80 mm; HR: 1.7, 95% CI: 1.0 - 2.9, P = 0.037), distant metastasis (vs no distant metastasis; HR: 1.9, 95% CI: 1.1 - 3.4, P = 0.019) were independent predictors of PFS.

Table 3.

Univariate and Multivariate Cox Regression Analysis of the Association Between Patient Characteristics and Progression-Free Survival (PFS)

Variables	Univariate			Multivariate
	HR	95% CI	P	HR	95% CI	P
ECOG
PS 0	Ref
PS 1	1.8	1.1 - 2.8	0.020
Child-Pugh score
A 5	Ref			Ref
A 6	2.2	1.4 - 3.7	0.002	2.0	1.2 - 3.4	0.009
B 7-8	9.3	2.7 - 32.3	0.000	4.5	1.2 - 17.1	0.029
ALBI score
Grade 1	Ref
Grade 2 – 3	1.6	1.0 - 2.7	0.045
AFP classification
<400 ng/ml	Ref
≥400 ng/ml	1.3	0.8 - 2.1	0.312
NLR
≤2.9	Ref			Ref
>2.9	1.6	1.0 - 2.7	0.047	1.6	0.9 - 2.8	0.082
Liver tumor size
<80 mm	Ref			Ref
≥80 mm	1.6	1.0 - 2.7	0.046	1.7	1.0 - 2.9	0.037
Portal vein thrombosis
No	Ref
Yes	1.1	0.7 - 1.7	0.781
Distant metastases
No	Ref			Ref
Yes	1.6	1.0 - 2.7	0.052	1.9	1.1 - 3.4	0.019

Abbreviations: CI, confidence interval; HR, hazards ratio; Ref, reference; ECOG, Eastern Cooperative Oncology Group; PS, performance status; ALBI, albumin-bilirubin; AFP, alpha-fetoprotein; NLR, neutrophil-to-lymphocyte ratio.

The boldness indicates statistical significance.

Predictive Factors Associated with Overall Survival (OS)

As shown in the above Table 4, univariate Cox regression analysis recorded the following factors: local therapy before first-line treatment, ECOG performance status, Child-Pugh score, ALBI grade, and neutrophil/lymphocyte ratio (NLR) before treatment were related to OS. Multivariate analysis showed that ECOG-PS 1 (vs PS 0: HR: 1.8, 95% CI: 1.0 - 3.3, P = 0.047), ALBI grade 2-3 (vs ALBI grade 1; HR: 2.0, 95% CI: 1.1 - 3.9, P = 0.033), NLR >2.9 (vs NLR ≤2.9; HR: 2.0, 95% CI: 1.1 - 3.6, P = 0.023) were independent predictors of OS.

Table 4.

Univariate and Multivariate Cox Regression Analysis of the Association Between Patient Characteristics and Overall Survival (OS)

Factors	Univariate			Multivariate
	HR	95% CI	P	HR	95% CI	P
Local therapy before first-line
Yes	Ref			Ref
No	1.9	1.0 - 3.5	0.047	1.7	0.9 - 3.3	0.085
ECOG performance status
PS 0	Ref			Ref
PS 1	1.8	1.0 - 3.2	0.045	1.8	1.0 - 3.3	0.047
Child-Pugh score
A 5 score	Ref
A 6 score	2.1	1.1 - 3.9	0.023
B 7-8 score	1.6	0.4 - 6.7	0.555
ALBI grade
Grade 1	Ref			Ref
Grade 2 – 3	2.2	1.2 - 4.2	0.017	2.0	1.1 - 3.9	0.033
AFP classification
<400 ng/ml	Ref
≥400 ng/ml	1.2	0.7 - 2.2	0.521
NLR
≤2.9	Ref			Ref
>2.9	2.1	1.2 - 3.8	0.011	2.0	1.1 - 3.6	0.023
Liver tumor size
<80 mm	Ref
≥80 mm	1.6	0.9 - 3.0	0.124
Portal vein thrombosis
No	Ref
Yes	1.2	0.7 - 2.2	0.478
Distant metastasis
No	Ref
Yes	1.0	0.6 - 1.9	0.950

Abbreviations: CI, confidence interval; HR, hazards ratio; Ref, reference; ECOG, Eastern Cooperative Oncology Group; PS, performance status; ALBI, albumin-bilirubin; AFP, alpha-fetoprotein; NLR, neutrophil-to-lymphocyte ratio.

The boldness indicates statistical significance.

Discussion

In our study, we evaluated efficacy of Atezolizumab plus Bevacizumab regimen for unresectable HCC patients with HBV infection. The results showed that mOS was 20 months, and mPFS was 4.3 months.

In the IMbrave150 trial, atezolizumab plus bevacizumab was the first-line standard treatment for unresectable HCC, reporting a median PFS of 6.9 months and OS of 19.2 months (HR 0.58, 95% CI 0.42-0.79) compared to sorafenib. ¹³ In the HBV subgroup of IMbrave150, the median PFS was 6.7 months and OS was 19.1 months. ¹⁵ Our study’s mOS (20 months) aligns closely with these results, suggesting comparable long-term benefits in HBV-related HCC. However, our mPFS (4.3 months) is notably shorter, which may reflect real-world challenges. Unlike IMbrave150’s controlled setting with strict inclusion criteria (eg, Child-Pugh A liver function), our study included 3.1% of patients with Child-Pugh B status, which is associated with reduced drug tolerance and poorer prognosis due to impaired liver function. Additionally, our cohort had a high tumor burden, with 61.5% of patients presenting with tumors ≥80 mm and 48.5% with portal vein thrombosis, both established predictors of poor PFS. ¹⁶ The presence of distant metastases in 37.1% of patients further likely contributed to accelerated disease progression, reducing PFS compared to the more selective IMbrave150 population.

Epidemiological characteristics related to HCC patients’ survival have also been researched and analysed.^17,18 To evaluate the impact of epidemiological characteristics on the efficacy of the Atezolizumab-Bevacizumab regimen in HCC patients, Federico Rossari et al conducted a study on 885 HCC patients from 42 centers in 5 countries in Europe and Asia (Italy, Germany, Portugal, Japan, and the Republic of Korea), of whom, 53.9% had hepatitis virus and 46.1% had not. Their outcomes were no difference in OS (15.9 months in viral group and 16.3 months in nonviral group). However, only 23.3% of this study population had chronic hepatitis B infection. ¹⁸ Similarly, Hatanaka et al (2023) conducted a study in 323 HCC patients with viral and non-viral infection using Atezolizumab-Bevacizumab in Japan, reporting a median PFS of 7.0 months in viral HCC (62.4% HCV, 37.6% HBV), higher than our 4.3 months. ¹⁹ This discrepancy may stem from the higher proportion of HCV-related HCC in their cohort, which is associated with better baseline liver function compared to HBV-related cases, as well as differences in healthcare access and tumor staging practices between Japan and Vietnam.

Comparing to other therapies, our findings highlight the advantages of atezolizumab plus bevacizumab over tyrosine kinase inhibitors like lenvatinib in HBV-related HCC. To compare the clinical action of lenvatinib in HBV-related HCC and HCV-related HCC, Xiaomi Li et al. conducted a study on 203 cancer patients in China. ²⁰ The results showed a median PFS of 6.1 months and OS of 14.9 months with lenvatinib in HBV-related HCC, with an objective response rate of 20.8% and disease control rate of 76.4%. ²⁰ While lenvatinib’s mPFS is higher than ours, the superior mOS in our study (20 months) underscores the durable benefits of combining immunotherapy with anti-angiogenesis, likely due to synergistic effects on the immunogenic tumor microenvironment and tumor vasculature. ²¹

The clinical importance of our findings extends beyond statistical metrics. The median PFS of 4.3 months, though lower than in controlled trials, represents a meaningful delay in disease progression for HBV-related HCC patients, a population with historically poor prognosis.^22,23 In real-world settings, extending PFS by even a few months allows for improved symptom management, potential eligibility for subsequent therapies, and better quality of life, particularly in HBV-endemic regions like Vietnam and Asian region. The median OS of 20 months is clinically significant, closely approaching IMbrave150’s viral subgroup (19.1 months in HBV group) and surpassing lenvatinib’s outcomes.^15,20 This suggested that atezolizumab plus bevacizumab offers substantial survival benefits, likely due to HBV-driven immune activation enhancing ICI responsiveness. For example, preclinical studies indicated that HBV infection upregulates PD-L1 expression, increasing sensitivity to atezolizumab. ²⁴ However, the shorter PFS compared to other studies underscores challenges in managing high tumor burden and advanced disease, necessitating strategies like multidisciplinary care to optimize outcomes.

Regarding survival predictors, our study found that liver function according to the Child-Pugh score, liver tumor size with a cut-off point of 8 cm, and distant metastasis, are independent survival predictors for PFS, whereas performance status, liver function according to the ALBI grade, and NLR are independent survival predictors for OS. Prognostic predictors when using the Atezolizumab-Bevacizumab regimen in HCC patients have also been mentioned in many studies but have not been mentioned specifically for patients with HBV. Tiago de Castro’s study conducted on 147 HCC patients in Germany and Austria using the Atezolizumab-Bevacizumab showed that ALBI and ECOG are 2 independent prognostic factors for OS. OS in patients with ALBI grade 1, 2, and 3 was 15 months, 8.6 months, and 3.2 months, respectively. OS in the PS 0 group was 15 months and PS ≥2 was 4.4 months. ²⁵ Regarding the neutrophil/lymphocyte ratio (NLR) in peripheral blood, it reflects the relationship between systemic inflammatory response and anti-cancer immune response. Several studies have demonstrated that a high NLR is associated with poor prognosis in cancer patients in general, and particularly in those with HCC.^26,27 However, the meaningful cut-off point of this index has not been consistent among studies. A study by Tada et al. showed that NLR ≤3 is associated with OS (hazard ratio: 3.369; 95% confidence interval, 1.024-11.080) when using the Atezolizumab-Bevacizumab for 249 patients in Japan. ²⁸ Young Eun Chon’s study conducted on 121 patients in Republic of Korea showed that NLR ≥2.5 was associated with poor prognosis of OS and PFS. ²⁹ Federico Rossari’s study also showed that factors such as NLR <3, ALBI grade 1 were independent factors associated with OS, but were not related to viral or nonviral hepatitis. ¹⁸ In our study, NLR ≤2.9 was associated with good prognosis of OS. Thus, the factors related to survival prediction in our study are quite similar to those in other studies on the HCC population in other countries.

Regarding the limitations, our study has several limitations that might impact its interpretation and generalizability. First, the modest sample size (n = 97) limited statistical power, potentially masking subtle subgroup differences and increasing the risk of type II errors, particularly for rare adverse events or predictors. Second, the retrospective data collection could generate selection and recall biases, especially in imaging assessments for PFS, which may overestimate or underestimate progression rates. Third, the geographical restriction to Northern Vietnam, where HBV is endemic, limits applicability to regions with different etiologies (eg, NASH-driven HCC in Western populations). This regional bias may explain our lower PFS compared to studies with mixed etiologies. Fourth, the short follow-up duration (median 11.1 months) restricts the assessment of long-term outcomes, such as durable responses or late recurrences, potentially underestimating OS benefits. Finally, the lack of a control group (eg, HBV-negative patients) due to the low prevalence of non-viral HCC in Vietnam hinders comparative efficacy analyses. These limitations collectively reduce the robustness of our findings and require cautious interpretation when extrapolating to broader populations. To address these limitations, future researchs should prioritize prospective, multicenter trials with larger, more diverse cohorts to enhance statistical power and generalizability across etiologies and regions.

Overall, our study proved that the Atezolizumab-Bevacizumab regimen in the first-line treatment of patients with unresectable HCC with HBV infection may help improve survival for patients in the real clinical settings. mOS was 20 months, which is a promising signal for this patient group with a reserved prognosis. Positive predictors for OS when using this regimen include performance status PS 0, liver function ALBI grade 1 and NLR ≤2.9. Hence, our findings provide real-world evidence that may help policy makers in developing treatment strategy in hepatitis B- endemic regions and support future researchs. In addition, future researchs investigating atezolizumab plus bevacizumab in specific subgroups, such as elderly patients (>75 years) or those with Child-Pugh B liver function, could clarify efficacy and safety profiles, strenthening real-world evidence. Finally, health economic analyses in resource-limited setting like Vietnam would evaluate cost-effectiveness, facilitating broader adoption of this regimen.

ORCID iDs

Nguyen Thi Dung https://orcid.org/0000-0001-5798-8636

Truong Thu Hien https://orcid.org/0009-0006-5766-0220

Pham Cam Phuong https://orcid.org/0000-0002-4342-0959

Ethical Considerations

The study protocol was approved by the Ethics Committee of Hanoi Medical University (Decision No. 1072/GCN-HMUIRB, dated February 27, 2024).

Consent to Participate

All prospective participants provided written informed consent prior to enrollment. For retrospective cases, informed consent was obtained through telephone contact with the patients or their family members (if patients died). Additionally, written approval for data usage was granted by the leaders of 3 hospitals. All patient information used in this study was fully de-identified prior to analysis in accordance with ethical standards, and no identifiable personal data were collected, stored, or reported.