Experience in improving treatment outcomes for childhood acute lymphoblastic leukemia: real-world results for a province in China, 2011–2020

Abstract

The present study aimed to investigate the real-world results of childhood acute lymphoblastic leukemia (cALL) cases in Fujian, China. The clinical data of 1414 patients with newly diagnosed cALL in Fujian were retrospectively analyzed. Patients were treated according to the Chinese Children Leukemia Group 2008 protocol (CCLG-ALL 2008 group) or Chinese Children’s Cancer Group 2015 protocol (CCCG-ALL 2015 group). Cumulative incidence of treatment abandonment (TA) at 5 years was 4.2% ± 0.6% and significantly associated with treatment period and risk stratification. The 5-OS and EFS were significantly higher in the CCCG-ALL 2015 group than in the CCLG-ALL 2008 group. Patients treated with CCCG-ALL 2015 from Fujian Medical Union Hospital had a significantly higher 4-year OS and EFS than did those from the other four hospitals. Real-world TA of cALL greatly decreased, and its long-term survival significantly increased in Fujian, which may be related to optimizing programs, multi-center collaboration, and improving treatment compliance.

Keywords:

• Leukemias, acute
• pediatric hematology/oncology
• long-term survival

Introduction

Acute lymphoblastic leukemia (ALL) is the most common childhood malignant neoplasm [1]. In developed countries, the 5-year event-free survival (EFS) and overall survival (OS) rates of childhood ALL (cALL) have dramatically improved to 85% and 90%, respectively [2–4]. However, the cure rates of cALL in developing countries still lag considerably behind developed countries, with the 5-year OS < 70% [5–7]. Such a remarkable survival gap can be mainly explained by higher rates of treatment-related mortality and relapse, due to inadequate supportive care infrastructure, lack of accurate risk stratification, and current treatment methods, such as tyrosine kinase inhibitors (TKIs), and chimeric antigen receptor (CAR)-T cell therapy. Another important reason is treatment abandonment, which is frequently related to economic difficulties, sociocultural factors, and inefficient medical care systems [8–10].

In Chinese Mainland, the age-standardized 5-year net survival for patients with cALL (0–14 years) was lower than 65% before 2014 according to Global surveillance of trends in cancer survival 2000–14 (CONCORD-3) reports [5]. In the first nationwide collaborative study in China, Chinese Children Leukemia Group (CCLG)-ALL 2008 study, the 5-year OS of patients with newly diagnosed cALL had improved remarkably to 85.3% [11]. Another nationwide collaborative trial in China, Chinese Children’s Cancer Group (CCCG)-ALL 2015 trial, reported that the 5-year OS of patients with newly diagnosed cALL had reached 90% [12]. However, patients who abandoned treatment and lack of compliance in those studies were excluded. Additionally, patients from hospitals in regions with relatively underdeveloped economic and medical conditions were not included. Thus, real-world cure rates for cALL may be lower than previously reported [11,12].

As Fujian is a middle-income region, and the New Rural Cooperative Medical Scheme (NRCMS) covered most costs for cALL treatment, most children with ALL could receive treatment after 2010. However, before 2015, Fujian Medical University Union Hospital was the only pediatric oncology unit with trained physicians and nurses in our province. Thus, patients outside Fuzhou abandoned treatment due to economic difficulties and transportation inconveniences. Before 2021, no patients receiving ALL treatment in Fujian were included in nationwide collaborative studies. The present study aimed to investigate the survival results of all patients with cALL, who were admitted to five hospitals in Fujian province from January 2011 to December 2020, and to summarize the real-world prognostic factors of cALL.

Material and methods

Patients and methods

Five centers in Fujian Province could diagnose and treat patients with newly diagnosed cALL before 2021. We searched the inpatient medical record systems of these five hospitals using the keywords ‘acute lymphoblastic leukemia’ and ‘age ≤ 14 years’ between January 2011 and December 2020. ALL was diagnosed based on the 2016 revision of the World Health Organization classification of lymphoid neoplasms [13]. Unified inclusion and exclusion criteria are available in Supplementary Table S1. Thirty-five patients were excluded: 12 with mature-B-ALL, 21 with mixed-phenotype acute leukemia, and four with trisomy-21 syndrome. A total of 1414 patients with previously untreated cALL were enrolled. Data collected and retrospectively analyzed included demographic data, central nervous system (CNS) and testicular involvement at diagnosis, MICM (morphological, flow cytometric, cytogenetic, and molecular biological) analyses results, treatment response, events (including relapse, death, abandonment, transfer, lost to follow up), and date of last follow-up.

Treatment

Of the 1414 patients, 20 patients abandoned treatment, four transferred outside Fujian, and two died of cerebral hemorrhage before chemotherapy. Of the 1388 patients who received chemotherapy, 968 received CCLG-ALL 2008 regimen [11], and 420 patients received CCCG-ALL 2015 regimen [12]. From January 2011 to December 2018, the patients of Fujian Medical University Union Hospital were treated with CCLG-ALL 2008 regimen. Because this hospital had become an observer unit for CCCG as of January 2019, the patients newly diagnosed between January 2019 and December 2020 were treated with the CCCG-ALL 2015 regimen which was supervised and quality controlled by the scientific committee of the collaborative group. Collaborative group experts provided prompt consultation for problem cases. The patients of Zhangzhou Affiliated Hospital of Fujian Medical University and First Affiliated Hospital of Xiamen University were treated with CCLG-ALL 2008 regimen. The patients of Quanzhou First Hospital Affiliated with Fujian Medical University and Nanping First Hospital of Fujian Province were treated with CCCG-ALL 2015. The treatment regimen of the CCLG-ALL 2008 was modified from the Berlin-Frankfurt-Munster (BFM) and Children’s Oncology Group (COG) protocol. Its risk-stratification system was based on cytogenetic subtypes, 7-day prednisone response, Day 15 (time point 1, TP1) and Day 33 (time point 2, TP2) bone marrow (BM) response, as previously described [11]. The treatment regimen of the CCCG-ALL 2015 was adapted from St Jude Children’s Research Hospital Total X V Study and Shanghai Children’s Medical Center (SCMC) ALL 2005 protocol. Its risk stratification system was based on cytogenetic subtypes, and Day 19 (TP1) and Day 46 (TP2) minimal residual disease (MRD), as previously described [12]. The comparison between CCLG-ALL 2008 and CCCG-ALL 2015 regimen and risk stratification criteria are shown in Supplementary Tables S2 and S3.

Definitions

Complete remission (CR) was defined as the absence of leukemia-related symptoms and signs, and <5% blasts morphologically in bone marrow. If no event occurred, the observation time was censored at the cutoff date for follow-up (31 December 2022). EFS was calculated from the date of diagnosis until the date of relapse, death, abandonment, or being alive at the last follow-up. OS was calculated from the date of diagnosis until the date of death or being alive at last follow-up. Treatment abandonment was defined as rejection/non-initiation of any treatment after diagnosis and failure to complete a potentially curative therapy program after treatment initiation in any hospitals [8]. Transfer was defined as transferred to hospitals outside of our province before completing a potentially curative therapy program. Patients with treatment abandonment without serious adverse events (SAEs) or lost to follow-up during CR were excluded upon last contact.

Statistical analysis

Between-group differences for categorical variables were compared by Pearson’s χ² test or Fisher’s exact test. Between-group differences for continuous variables were compared by Wilcoxon-Mann-Whitney tests. The OS and EFS rates were calculated using the Kaplan-Meier method; groups were compared by log-rank test. Univariate analyses were performed by Cox proportional-hazard regression model; factors that were significantly different were further performed by multivariate analyses. A two-sided p-value of <0.05 was considered statistically significant. All statistical analyses were performed with SPSS 25.0 (SPSS Inc., Chicago, Illinois, USA) and GraphPad Prism software 7.0 (GraphPad Software, San Diego, California, USA).

Results

Patients’ characteristics

Of the 1414 patients with cALL (841 boys; 573 girls), the median age was 51 (4–168) months and 26 cases were infant ALL. The white blood cell counts of 331 patients at diagnosis were higher than 50 × 10⁹/L. Thirty-two and 10 cases included CNS and testicular involvement, respectively. Flow cytometry analysis was detected in all patients: 1286 B-lineage and 128 T-lineage, respectively. Patients’ characteristics are summarized in Supplementary Table S4.

Events

The flow diagram of events is shown in Supplementary Fig S1.

Of the 1414 patients, 56 abandoned treatment, including 20 who abandoned before chemotherapy and 36 who abandoned during chemotherapy without SAEs or relapse. The cumulative incidence of treatment abandonment at 5 years was 4.2% ± 0.6% (Figure 1A). Patients were divided into three groups according to treatment period (2011–2015, 2016–2018, 2019–2020). Abandonment rates were 5.3%, 4.7%, and 1.5%, respectively. Treatment abandonment was significantly associated with treatment period (before or after 2019) (5.0% vs. 1.5%, p = 0.002), risk stratification (high risk or non-high risk) (7.0% vs. 3.0%, p = 0.001), especially for the patients with BCR-ABL1 (12.7% vs. 3.6%, p = 0.000), and MLL-rearrangement (10.4% vs. 3.7%, p = 0.020). The most common reason for treatment abandonment was economic difficulties (85.7%); other reasons included belief that ALL was incurable (7.1%), severe adverse effects of chemotherapy (5.4%), and religion (1.8%).

Figure 1. The cumulative incidence of treatment abandonment and relapse of the patients with childhood acute lymphoblastic leukemia (cALL). (A) The cumulative incidence of treatment abandonment of the 1414 patients with cALL; (B) The cumulative incidence of relapse (CIR) of the 1252 treatment-compliant patients (excluding 56 abandonment, 33 transfers, and 73 died of chemotherapy-related complications).

Thirty-three of the 1414 patients transferred to hospitals outside of our province before relapse. Reasons for transfer included poor treatment response (30.3%), severe adverse effects of chemotherapy (24.2%), receiving CART or HSCT in other hospitals (24.2%), patient’s intention (12.1%), and returning to the registered residence for treatment (9.1%).

Of the 1325 patients receiving chemotherapy, 73 died of chemotherapy-related complications: 78.1% died of infectious complications, and 21.9% died of noninfectious complications; 58.9% died during induction, and 41.1% died after induction.

Of the 1252 patients who complied with treatment, 140 relapsed. The cumulative incidence of relapse (CIR) at 5 years was 12.3% ± 1.0% (Figure 1B). The median time of relapse was 18.7 (range 2.4–97.2) months: 10.7%, <6 months; 37.1%, 6–18 months; 26.4%, 18–36 months; 25.7%, ≥36 months. After relapse, 47.1% abandoned therapy, 15.7% received only chemotherapy, 7.1% received only CART, 20.7% underwent HSCT, and 9.3% underwent HSCT after CART.

Survival and prognostic analysis

At a median follow-up of 54.6 (0.1–142.9) months, the 5- and 10-year OSs for the whole group were 83.0% ± 1.0% and 80.9% ± 1.2%, respectively (Figure 2A); the 5- and 10-years EFSs for the whole group were 77.7% ± 1.2% and 76.0% ± 1.3%, respectively (Figure 2B). Excluding 89 patients who abandoned treatment and transferred, estimated OSs at 5- and 10-years were 86.9% ± 1.0% and 84.8% ± 1.2%, respectively (Figure 2C); estimated EFSs at 5- and 10-years were 82.8% ± 1.1% and 81.6% ± 1.2%, respectively (Figure 2D).

Figure 2. The overall survival (OS) and event-free survival (EFS) of the patients with childhood acute lymphoblastic leukemia (cALL). (A) The OS of the 1414 patients with cALL; (B) The EFS of the 1414 patients with cALL; (C) The OS of the 1252 treatment-compliant patients (excluding 56 abandonments and 33 transfers); (D) The OS of the 1252 treatment-compliant patients (excluding 56 abandonments and 33 transfers).

Univariate Cox analysis results are shown in Supplementary Table S5. Age (<12 or ≥120 months), WBC ≥ 50 × 10⁹/L at diagnosis, T-ALL, BCR-ABL1, MLL-rearrangement were significant factors for poorer EFS and OS (all HR > 1 and p < 0.05). In contrast, ETV6-RUNX1 and hyperdiploid were favorable predictors for better EFS (HR = 0.578, p = 0.025) and OS (HR = 0.265, p = 0.001).

Multivariate Cox analysis results are shown in Supplementary Table S6. WBC ≥ 50 × 10⁹/L at diagnosis, T-ALL, BCR-ABL1, and MLL-rearrangement were independent risk predictors for poorer OS and EFS (all HR > 1 and p < 0.05), while hyperdiploid significantly affected EFS (HR = 0.431, p = 0.022) but not OS (HR = 0.604, p = 0.074).

According to treatment period, the patients were divided into three groups:2011–2015, 2016–2018, 2019–2020. The clinical characteristics of the there groups were not significantly different (all p ≥ 0.05) (Supplementary Table S7). The estimated OS and EFS at 4 years during 2011–2015 and 2016–2018 were similar (OS: 81.5% ± 1.9% vs. 81.5% ± 1.6%, p = 0.923; EFS: 77.7% ± 2.0% vs. 75.7% ± 1.8%, p = 0.624), but significantly lower than those during 2019–2020 (OS: 90.0% ± 1.5%, EFS: 86.3% ± 1.8%, p = 0.002 and <0.001, respectively) (Figure 3A, B).

Figure 3. The comparison of overall survival (OS) and event-free survival (EFS) of the patients with childhood acute lymphoblastic leukemia (cALL) according to treatment period (2011–2015, 2016–2018, 2019–2020). A: The comparison of OS of the patients with cALL according to treatment period; B: The comparison of EFS of the patients with cALL according to treatment period.

The correlation between survival probability and blast levels by morphological evaluation and MRD analysis at TP1 was available for 1313 patients and 1123 patients, respectively. We stratified the patients into three groups according to blast% by morphological evaluation at TP1: <5%, 5–25%, and ≥25%. The 5-year OS of the three groups differed significantly: 87.8% ± 1.0%, 72.4% ± 9.4%, and 56.1% ± 9.5%, respectively, (p < 0.001) (Figure 4A). The 5-year EFS of the three groups also differed significantly: 81.9% ± 1.2%, 65.2% ± 6.2%, and 37.4% ± 9.5%, respectively, (p < 0.001) (Figure 4B). Patients were stratified into three groups according to MRD levels at TP1: <0.1%, 0.1–1.0%, and ≥1.0%. The 5-year OS of the MRD <0.1% and 0.1–1.0% groups was similar: 87.8% ± 2.2% and 92.4% ± 1.2%, respectively (p = 0.137), which were significantly higher than the MRD ≥1.0% groups (81.2% ± 2.1%, p < 0.001, Figure 4C). The 5-year EFS of the three groups also differed significantly: 87.7% ± 1.6%, 81.2% ± 2.7%, and 72.7% ± 2.5%, respectively, (p < 0.001) (Figure 4D).

Figure 4. The comparison of overall survival (OS) and event-free survival (EFS) of the patients with childhood acute lymphoblastic leukemia (cALL) according to early treatment response at time point 1 (TP1). (A) The comparison of OS of patients with cALL according to blast percent by morphological evaluation at TP1; (B) The comparison of EFS of patients with cALL according to blast percent by morphological evaluation at TP1; (C) The comparison of OS of patients with cALL according to MRD levels at TP1; (D) The comparison of EFS of patients with cALL according to MRD levels at TP1. TP1: was defined as Day 15 during induction for the patients who received CCLG-ALL 2008 regimen and Day 19 during induction for the patients who received CCCG-ALL 2015 regimen.

The correlation between survival probability and blast levels by morphological evaluation and MRD analysis at TP2 was available for 1335 patients and 1288 patients, respectively. We stratified the patients into two groups according to blast% by morphological evaluation at TP2: <5%, and ≥5%. The 5-year OS and EFS of the patients with <5% blast at TP2 were all significantly higher than patients with ≥5% (OS: 88.5% ± 1.0% vs. 60.0% ± 6.3%, p < 0.001; EFS: 82.4% ± 1.2% vs. 47.7% ± 6.3%, p < 0.001) (Figure 5A, B). We stratified the patients into three groups by MRD levels at TP2: <0.01%, 0.01–0.1%, and ≥0.1%. The 5-year OS of the three groups also differed significantly: 90.9% ± 1.0%, 81.5% ± 3.8%, and 65.3% ± 5.4%, respectively, (p < 0.001) (Figure 5C). The 5-year EFS of the three groups also differed significantly: 85.4% ± 1.2%, 76.2% ± 4.1%, and 53.3% ± 5.6%, respectively, (p < 0.001) (Figure 5D).

Figure 5. The comparison of overall survival (OS) and event-free survival (EFS) of the patients with childhood acute lymphoblastic leukemia (cALL) according to early treatment response at time point 2 (TP2). (A) The comparison of OS of patients with cALL according to blast percent by morphological evaluation at TP2; (B) The comparison of EFS of patients with cALL according to blast percent by morphological evaluation at TP2; (C) The comparison of OS of patients with cALL according to MRD levels at TP2; (D) The comparison of EFS of patients with cALL according to MRD levels at TP2. TP2: was defined as Day 33 during induction for the patients who received CCLG-ALL 2008 regimen and Day 46 during induction for the patients who received CCCG-ALL 2015 regimen.

Comparison of CCLG-ALL 2008 group and CCCG-ALL 2015 group

Between 2016 and 2020, 420 patients were diagnosed and treated with CCCG-ALL 2015 regimen (CCCG-ALL 2015 group) while 542 patients were treated with CCLG-ALL 2008 regimen (CCLG-ALL 2008 group). The clinical characteristics of the two groups were not significantly different (all p ≥ 0.05) (Supplementary Table S9).

The TRM rate of the CCLG-ALL group was significantly higher than the CCCG-ALL 2015 group (6.3% vs. 3.3%, p = 0.049). The 5-year EFS and OS of the CCCG-ALL 2015 group were significantly higher than CCLG-ALL 2008 group (OS: 88.3% ± 2.1% vs. 83.0% ± 1.6%, p = 0.003; EFS: 83.7% ± 3.3% vs. 76.4% ± 1.9%, p < 0.001) (Figure 6A, B). Further subgroup analysis was performed based on different hospitals: Fujian Medical Union Hospital and the other four hospitals (combined due to the small sample). The clinical characteristics of the two groups were not significantly different (all p ≥ 0.05) (Supplementary Table S8). The patients treated with CCLG-ALL 2008 from Fujian Medical Union Hospital had a similar 5-year OS and EFS to the patients treated with CCLG-ALL 2008 from the other four hospitals (OS: 83.9% ± 1.8% vs. 79.1% ± 3.9%, p = 0.254; EFS: 78.2% ± 2.0% vs. 68.4% ± 4.6%, p = 0.104) (Supplementary Fig S2A, S2B). In contrast, patients treated with CCCG-ALL 2015 from Fujian Medical Union Hospital had a significantly higher 4-year OS and EFS (because the follow-up time is too short) than patients from the other four hospitals (OS: 89.9% ± 3.3% vs. 84.2% ± 3.3%, p = 0.019; EFS: 89.1% ± 2.1% vs. 82.3% ± 3.4%, p = 0.046) (Figure 6C, 6D).

Figure 6. The comparison of overall survival (OS) and event-free survival (EFS) of patients with childhood acute lymphoblastic leukemia (cALL) according to different treatment regimens (CCLG-ALL 2008 group and CCCG-ALL 2015 group). (A) The comparison of OS of patients with cALL according to different treatment regimens; (B) The comparison of EFS of patients with cALL according to different treatment regimens; (C) The comparison of OS of cALL patients treated with CCCG-ALL 2015 from different hospitals (Fujian Medical Union Hospital and other four hospitals); (D) The comparison of EFS of cALL patients treated with CCCG-ALL 2015 from different hospitals (Fujian Medical Union Hospital and other four hospitals).

Discussion

This retrospective analysis of clinical data for all newly diagnosed cALL hospitalized in five centers in Fujian from 2011 to 2020 revealed several important biological features. The 1.8% incidence of infant ALL was similar to that (0.7% to 1.1%) in previous reports from China [11,12,14], but was lower than that in reports from Western countries of 2% to 5% [15,16]. This difference may be due to the high abandonment rate among cases of infant ALL. Previous Western studies report that KMT2A (MLL) rearrangement (KMT2A-r) occurs in 5–6% of cALL and approximately 75% of infants with B-ALL [16,17]. However, herein KMT2A-r occurred in only 3.4% of cALL cases, similar to some studies from China reporting 2% to 3% [11,12,14]. This result may be due to a lower incidence of infant ALL in this study. Compared to Western counties [18–20], other important biological features identified in our study included a significant decrease in the incidence of T-ALL (9.0% vs. 15.0%), ETV6-RUNX1 (17.9% vs. 25%), and high hyperdiploid (12.2% vs. 25%), but a slight increase in TCF3-PBX1 (5.9% vs. 4.0%). However, the main clinical and biological features of cALL in our study are similar to most previous studies from both Western counties and China [4,21–24]. Therefore, the patients enrolled in our study are comparable to those in previous studies.

The present results also indicated that the 5-year OS and EFS were lower than 85% and 80%, respectively, indicating an appreciable gap compared to domestic countries [2,23]. This result may be related to the following three factors. First, the cumulative incidence of abandonment of newly diagnosed and relapsed patients in this study was nearly 5% and 50%, respectively, which are higher than those in studies conducted in developed countries [25]. Second, the TRM rate in our study was 5.5%, which is significantly higher than that in previous reports (∼2–4%) [26,27]. Third, our data lack next-generation whole-genome and transcriptome sequencing for more precise risk stratification and therapeutic target drugs.

Fortunately, the EFS and OS of cALL patients in our province had gradually improved since 2013, especially from 2019 to 2020. One important reason for the improvement in EFS and OS is that fewer patients abandoned treatment. Abandonment due to cost of cALL mainly includes two parts: cost of in-hospital treatment and cost to families. Since 2010, when NRCMS began to cover most costs for cALL, the rate of abandonment significantly decreased [8,10]. However, some therapies, such as expensive antibiotics that are not covered by NRCMS, still cause financially at-risk patients to abandon treatment. Before 2015, Fujian Medical Union Hospital was the only dedicated pediatric leukemia center in Fujian; therefore, patients outside Fuzhou experienced higher personal costs. Four hospitals in Fuzhou had pediatric oncology units with trained physicians and nurses since 2015, thus, some financially at-risk patients or those with transportation difficulties could be treated locally. In addition, Fujian Anti-poverty Charity Association and Children’s Hope Foundation (Fuzhou Xiaojia), founded by social philanthropists, provided free accommodation for patients from other areas and provided funding for many impoverished low-risk patients and patients who encountered serious complications during the remission period, thus, preventing many patients from abandoning. In 2019, the government required medical insurance to pay for hospitals based on Diagnosis Related Groups, further reducing hospital costs. The Fujian Golden Ribbon Public Welfare Platform, jointly established by patients, volunteers, and doctors, was established and provides funding and spiritual assistance to patients and their families. Therefore, we believe that the continuous optimization of government medical insurance, the intervention of social charity funds, and the localization of treatment may help reduce abandonment rates.

Another reason for the survival improvement is the reduced TRM rate, which may be mainly related to the optimization of treatment regimens. The incidence of TRM in contemporary ALL trials is 2–4%, mostly due to infections [27,28]. Intensive chemotherapy and lack of neutrophil surge are significantly associated with infection-related complications [29]. A study from Japan revealed that most fatal infections occurred in the third week of induction [30]. Our previous study reported that TRM rates and SAEs, especially the infection-related SAEs of CCLG-ALL 2008 were significantly higher than that of CCCG-ALL 2015 [31]. In this larger sample study, we further confirmed that the CCLG-ALL 2008 group’s TRM rate was significantly higher than that of the CCCG-ALL 2015 group. Compared to the CCLG-ALL 2008 regimen, the CCCG-ALL 2015 regimen reduced the intensity of chemotherapy in a single course of treatment. For instance, during induction, patients require 2–4 doses of daunorubicin in CCLG-ALL 2008 group, while the patients require 1–2 doses of daunorubicin during induction in the CCCG-ALL 2015 group; during consolidation chemotherapy, high-risk patients required chemotherapy of high-dose methotrexate combined with multiple drugs in the CCLG-ALL 2008 group, while patients only received high-dose methotrexate in the CCCG-ALL 2015 group. The reduced intensity of chemotherapy significantly shortens the duration of neutropenia, thus, reducing the incidence of infection-related complications.

Our results indicate that patients treated with CCCG-ALL 2015 from Fujian Medical Union Hospital had significantly higher survival rates than patients treated with CCCG-ALL 2015 from the other four hospitals. Although the follow-up time was relatively short, we speculated that participation in a collaborative study may be responsible for this result. Collaborative study has been recognized as an effective way to improve the survival rate of cALL [4]. Collaborative study of cALL began relatively late in China. The CCLG-ALL 2008 trial, designed to improve survival rates for cALL, was the first nationwide collaborative study in China [11]. Since 2015, the CCCG-ALL 2015 trial was the largest nationwide collaborative study in China [12]. Participants with cALL in collaborative studies experienced an improvement to a 5-year OS rate of 85%–90% [11,12]. However, the four hospitals in Fujian other than Fujian Medical University Union Hospital, similar to other regions with relatively underdeveloped economic and medical conditions, did not participate in any collaborative study. Therefore, our future recommendations include establishing collaborations with local clinicians, national and international disease experts, stratifying risk more precisely by new technologies such as RNA sequencing, developing protocols that are more suitable to local conditions, and providing prompt consultation for problem cases.

There are several limitations in this study, including, but not limited to, its retrospective nature, some outpatients who abandoned treatment were excluded, the follow-up time of patients treated with CCLG-ALL 2015 was short, treatment delays were experienced by some patients, and it lacked MRD evaluation. Although we provided unified training for data collection, the data collection and quality control remained challenging.

In conclusion, the treatment abandonment of cALL has greatly decreased and long-term survival of patients with cALL has significantly increased in Fujian province since 2013. However, a gap remains compared to developed countries, which was not unexpected, as Fujian remains a developing area. We observed that the outcome of cALL can be improved further by optimizing treatment programs, multi-center collaborations, and improving treatment compliance. In the next decade, we will continue to adapt efficient treatment strategies for use in the current situation in Fujian to improve treatment for cALL.

Ethical approval

This study was approved by the institutional ethics committees of all five participating hospitals: Fujian Medical University Union Hospital, the First Affiliated Hospital of Xiamen University, Zhangzhou Affiliated Hospital of Fujian Medical University, Quanzhou First Hospital Affiliated to Fujian Medical University, and Nanping First Hospital of Fujian Province. Because the study itself did not involve protocol adjustments or patient privacy, the hospital ethics committee agreed to waive informed consent.

Author contributions

Yongzhi Zheng, Jiazheng Li, Jianda Hu, Shaohua Le conceived and designed the experiments. Yongzhi Zheng and Jiazheng Li performed the experiments. Yongzhi Zheng, Shaohua Le, Hao Zheng, Xueling Hua, Zaisheng Chen, Hong Wen, Kaizhi Weng, Shuquan Zhuang, Xingguo Wu collected clinical data. Yongzhi Zheng and Jiazheng Li analyzed and interpreted the data. Yongzhi Zheng and Jiazheng Li wrote the manuscript. Shaohua Le and Jianda Hu critically revised manuscript. All authors read and approved the manuscript.

Acknowledgments

We would like to express our deepest gratitude to patients who donated samples for research purposes. We would like to express our sincere thanks to the doctors of the cooperative units for providing the clinical data. We would like to thank Editage (www.editage.cn) for English language editing.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Data availability statement

The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.

Additional information

Funding

This work was supported by the Young and Middle-aged Scientific Research Major Project of Fujian Provincial Health Commission [No. 2022ZQNZD012], the National Key Clinical Specialty Discipline Construction Program [2021-76] and Fujian Provincial Clinical Research Center for Hematological Malignancies [2020Y2006].