Despite being the most prevalent cancer among children, over 90% of pediatric acute lymphoblastic leukemia (ALL) cases are cured in most developed countries^.1 This has largely been attributed to the introduction of risk stratification protocols and risk-directed therapies.² However, there remains heterogeneity in treatment responses within risk groups which suggests the need for risk stratification to be further refined.² The predictive potential of CD9, a tetraspanin family protein that regulates cell motility, adhesion and signaling, in survival outcomes of patients with childhood B-cell precursor acute lymphoblastic leukemia (BCP-ALL) was previously highlighted in a local single-center investigational study.³ Thereafter, researchers from The Chinese University of Hong Kong (CUHK) and clinicians in the Hong Kong Children’s Hospital conducted a nationwide, multicenter, retrospective study to validate clinical performance of CD9 as a prognostic biomarker in childhood ALL.²

The retrospective cohort study included pediatric patients diagnosed with ALL under the age of 18 who were enrolled in the Chinese Children Cancer Group (CCCG)-ALL-2015 clinical trial between 2015 and 2019 and had CD9 data available.² The analysis was conducted based on data from 3,395 B-cell acute lymphoblastic leukemia (B-ALL) patients and 386 T-cell acute lymphoblastic leukemia (T-ALL) patients, where CD9 positive (CD9+) rates (defined as the presence of ≥20% CD9+ blasts) were 88.5% and 27.2% for the 2 respective patient groups.² The 5-year event-free survival (EFS) rate and cumulative incidence of relapse (CIR) served as the primary endpoints of this study, while the secondary endpoints focused on evaluating the correlation between CD9 positivity and known risk factors for events such as patients’ age, gender and white blood cell (WBC) counts.²

Upon inspecting the results, CD9 positivity was found to be associated with a significantly inferior 5-year EFS for B-ALL patients compared to their CD9-negative (CD9-) counterparts (82.1% vs. 89.3%; p=0.001).² Such association between CD9+ and inferior 5-year EFS was especially pronounced among patients with day 19 minimal residual disease (MRD) (77.8% vs. 86.6%; p=0.031) and patients who were classified as intermediate or high-risk (72.8%vs. 83.2%; p=0.028).2 Furthermore, B-ALL patients who were CD9+ had a significantly higher CIR (15.5% vs. 7.8%; p<0.001), which was observable both in CD9+ patients with high (30.3%vs. 9.7%; p=0.007) and low (20.1% vs.10.8%; p=0.022) day 19 MRD, as well as in intermediate or high-risk patients (23.1% vs. 10.2%; p=0.003).² In contrast, no statistical significance in 5-year EFS and CIR between CD9+ and CD9- patients was observed in the T-ALL group (p=0.85 for 5-year EFS; p=0.58 for CIR).²

Subsequent analysis of the B-ALL group revealed that CD9 positivity was associated with high initial WBC count of >50x109/L, hyperdiploidy and TCF3::PBX1 (p<0.001 for all), whilst CD9 negativity was associated with ETV6::RUNX1 (p<0.001).² Based on these parameters, a higher proportion of CD9+ patients were classified as having an intermediate risk (p=0.002) whilst a lower proportion were classified as having a low risk of relapse (p=0.003).² In addition, results from the multivariate analysis affirmed CD9 positivity as an independent risk factor for events (HR=1.917; 95% CI: 1.313-2.798; p=0.001) and relapse (HR=2.208; 95% CI: 1.435-3.398; p<0.001), solidifying CD9’s performance as a prognostic biomarker.²

To conclude, the findings of this nationwide multicenter study established the prognostic value of CD9 among children with ALL, where CD9 positivity can be considered an independent predictor of relapse in patients with B-ALL, particularly in those with intermediate or high-risk diseases, positive MRD status or specific cytogenetic backgrounds.³ Thus, integrating this prognostic biomarker into the diagnostic immunophenotyping panel may help refine clinical risk stratification, thereby informing more proper management of childhood ALL.^2,3

Question 1: How prevalent is childhood ALL in Hong Kong?

Prof. Leung: Globally, the prevalence of childhood ALL is approximately 40-50 cases per 1,000,000 children which is largely similar between countries and regions. In Hong Kong, roughly 50-60 children are newly diagnosed with childhood ALL annually and are treated in the Hong Kong Children’s Hospital.

Question 2: How well is childhood ALL currently managed in Hong Kong? Can treatment strategies be improved?

Prof. Leung: While treatment options for ALL have remained largely unchanged, treatment outcomes have seen remarkable improvement with the overall cure rate currently exceeding 90% in Hong Kong and most developed countries. This success can be largely attributed to risk-directed therapies. Risk stratification has mostly been conducted based on factors such as patients’ age, WBC count, cytogenetic risks and MRD responses. However, some patients continue to experience relapses that are unexpected in their respective risk groups. This has indicated the need for further refinement in risk stratification such that treatments can be appropriately intensified or de-escalated.

Question 3: What challenges can you foresee hindering CD9’s utility as a prognostic biomarker in childhood ALL?

Prof. Leung: In our study, CD9 has established itself as a remarkable predictor for disease outcomes for B-ALL, exuding promising potential as an addition to the current diagnostic immunophenotyping panel. Nevertheless, the protocol for CD9 detection under flow cytometry requires standardization. In particular, the antibody fluorescence tag for CD9 detection should be standardized, as different fluorochromes exhibit different signals to accurately determine CD9 positivity. Moreover, the clinical benefit after the addition of CD9 for refining risk stratification will have to be formally tested in clinical trials.