DISCUSSION
Non-Hodgkin lymphoma is a heterogenous disease that has been reported to be consistently increasing in many parts of the world. Amongst the NHL, DLBCL is one of the most frequent types of aggressive NHL that accounts for 41% of the newly diagnosed adult NHL in Asia (14, 16). Although CR was achievable in 60% to 80% of patients, the durable remission was only seen in 35-45% of patients (17, 18). With AHSCT, the OS has improved, especially in RR-NHL (5, 8, 19, 20). AHSCT is now widely used as standard therapy in relapsed NHL (8) and most guidelines recommended it as consolidation following salvage chemotherapy (9, 21).
Several studies have reported good outcomes following HDT-ASHCT (20, 22). In this study, the three-year OS and EFS of 70.7% and 62.1% respectively is comparable to that of other centres worldwide, where the OS rate ranges from 68-74%, and the progression-free survival rate of 60-69% (5, 19). The low transplant-related mortality of AHSCT, which is comparable to other centres (16, 26), implied that supportive care in our centres was non-inferior.
In this study, AHSCT at CR expectedly yielded better OS and EFS when compared to those who were transplanted in PR and is consistent with other studies (23, 24). Patients who were transplanted in CR1 appeared to have better OS and EFS and this is especially evident in high-risk patients who were transplanted in CR1. Our findings were consistent to what was reported by Caballero et al. (2003), Nakaya et al. (2017), and Zhao et al. (2017). Similarly, a study by Kansai Medical University Hospital showed that upfront AHSCT provided better outcomes compared to those who did not undergo ASHCT (16), and this is further supported by a meta-analysis which found that high risk patients may benefit from upfront AHSCT (4). However, appropriate timing in transplantation remains debatable. Its role in the upfront setting remains controversial.
IPI scoring had been used as a clinical tool for risk stratification of patients with DLBCL and is considered valuable as a prognosis indicator in aggressive lymphoma (1, 28, 29).Through the revised IPI score, patients with high-risk IPI with DLBCL continue to have sub-optimal outcomes with a predicted five-year survival of 50–55% (3). Accurate molecular classification is extremely important for precision medicine in malignant lymphomas. Increasing, molecular and genotyping cell of origin (COO) are methods being used but limited as needs cost for Next-Generation Sequencing (NGS). In the early 2000, a study demonstrated that DLBCL could be differentiated based on gene expression profiling methods (GEP) into distinct molecular subtypes through their cell of origin (COO) (33), coupled with many other studies confirming the prognostic significance for molecular subtyping of COO subtypes of ABC DLBCLs having a poorer PFS (1,33 – 36) . In this study, IPI was not found to be predictive and this may be due to the relatively small number of DLBCL in this cohort, with a significant proportion of lymphomas were not classifiable. This raised the issue of the importance of the mutation profiling besides the possible downfall when using the clinical prognostic score alone in malignant lymphomas.
The limitations of this study are the sample size is relatively small and there is a relatively significant proportion of lymphoma were not classified according to the new WHO classification. The follow up period is also relatively short.
Despite these limitations, our study illustrates the outcomes of HDT and AHSCT in NHL patients in a resource limited country with a comparable TRM. Due to the higher incidence of aggressive lymphoma in Asian population, there is a need for Asian countries to construct multicentre databases to spur the kind of research demonstrated in developed countries, to find the most feasible treatment modalities such as upfront AHSCT which may be a cost-effective strategy to reduce the risk of relapse. Moving forward, we recommend larger studies in randomized and selected cohort with longer duration of follow up to be conducted to validate our findings.