Diffuse large B-cell lymphoma (DLBCL) is the most common type of lymphoma and accounts for approximately one-third of non-Hodgkin’s lymphoma [1]. The age-standardized rate (per 100,000) for male and female in Taiwan was only 3.44 and 2.54, respectively [1]. The causes of DLBCL are not well understood. The DLBCL can arise in any part of the body and is often an aggressive type of malignancy. Although it usually arises from the normal B lymphocytes, it can also represent a malignant transformation from other types of B-cell lymphoma. Typically, the first sign of this disease is the rapidly growing tissue or mass sometimes associated with the B symptoms (e.g., fever, weight loss, and night sweats). The diagnosis of DLBCL is made pathologically of the biopsied tissue. There are several subtypes of DLBCL identified which differ in their clinical presentations, symptoms, aggressiveness, morphology, immunophenotypes, gene expression, and the prognosis [2]. Because of its aggressiveness with complex outcomes due to the heterogeneous entity, complete pretreatment evaluation including prognostic classification is important for the disease management.
In the past 20 years, the international prognostic index (IPI) has been one of the most useful tools to evaluate the prognosis in patients with aggressive NHL including DLBCL. In the evaluation of IPI, one point is assigned for each of the following risk factors: (1) age over 60 years, (2) Stage III or IV disease, (3) elevated level of serum lactate dehydrogenase (LDH), (4) Eastern Cooperative Oncology Group (ECOG) performance status of 2–4, and (5) more than 1 extra-nodal site involvement of disease (i.e., site other than lymph nodes, spleen, thymus, and the pharyngeal lymphatic ring). The 5-year overall survival (OS) for low risk (IPI: 0–1), low-intermediate risk (IPI: 2), high-intermediate risk (IPI: 3), and high risk (IPI: 4–5) is 73%, 51%, 43%, and 26%, respectively [3].
Since the addition of rituximab (an immunoglobulin G1 monoclonal antibody against B-lymphocyte antigen CD20) into the first-line chemotherapy (i.e., cyclophosphamide, doxorubicine, vincristine and prednisone) (R-CHOP), the 5-year survival rate of DLBCL has been improved [4]. However, patients with the same IPI score may still suffer from different outcomes either due to early relapse or refractory disease. Great efforts have been made to improve the evaluation models for risk stratification [5–8], and more reliable prognostic predictors are pressingly needed to differentiate patients who are more likely to have poorer outcome [9]. Further information gained from different imaging modalities to build a more reliable prediction model for clinical practice is helpful and valuable.
As stated above, pretreatment staging is crucial. The fluorine-18 fluorodeoxyglucose (FDG), a glucose analog, can be used to measure the degree of glucose utilization. The uptake degree of FDG detected by positron emission tomography/computed tomography (PET/CT) represents the tissue metabolism on the whole-body and functional images. The FDG PET/CT has been widely used in pretreatment staging of disease and assessment of treatment response for patients diagnosed with DLBCL [10–12]. It has been reported in the literature that there are a variety of quantitative parameters deriving from image which have potential utility to predict prognosis or treatment outcome. The standardized uptake value (SUV) is the most commonly used parameter in FDG PET/CT and has been proved to be a significant prognostic predictor in DLBCL [13–14]. The percentage change in maximal SUV (SUVmax) between initial and delayed phase images is defined as the retention index (RI) which is found to have significant prognostic potential to predict overall survival (OS) in patient with DLBCL [15]. Beyond SUV and RI, the total metabolic tumor volume (MTVsum) has also been shown to be a predictor for survival outcome in many previous studies [16–19]. It was also reported that an elevated MTVsum, independent from IPI, is a predictor for shorter progression-free survival (PFS) and OS in patients with DLBCL [20].
Learning is the process of behavior improvement over time via discovering new information. If the referred process is achieved by machine rather than human brain, it is called machine learning. The experience acquiring from the existing examples helps to find the optimal solution for coming problems in the machine learning process. Due to rapid accumulation of larger and larger raw data, the traditional methods cannot handle well and the big data concept emergences over time with the development of information technologies. Machine learning in which algorithms were used by the computers with a certain order when performing operations is a subset of artificial intelligence. Based on the training data, machine learning algorithms can search for optimal connection weights of a prespecified neural network model in order to make decisions and predictions. Some machine learning models, in which sophisticated mechanism such as high-order and non-linear interactions between predictors and the responses were used, have shown ability to improve overall clinical prediction in various conditions [21–23].
Nonetheless, due to the heterogenous entity of DLBCL, the prognostic parameters of DLBCL, the total effects of these parameters, and individual weight of each parameter remain to be an issue deserving further research. Previous studies have reported that machine learning algorithms, using either molecular profiling data or combined clinical and genetic data, helped to do disease classification, diagnosis, and prognosis prediction [24–29]. However, there are few reports, in which using modern machine learning models and incorporating FDG PET/CT metabolic imaging parameters, to perform the outcome prediction in DLBCL.
Therefore, this study aims to develop the logistic and neural network models to predict DLBCL clinical outcome based on PFS and OS. Both clinical and metabolic parameters from FDG PET/CT scans were used as predictors. The performance of the models is also compared.