Development and validation of a machine learning model to predict time to renal replacement therapy in patients with chronic kidney disease.

BACKGROUND: Predicting time to renal replacement therapy (RRT) is important in patients at high risk for end-stage kidney disease. We developed and validated machine learning models for predicting the time to RRT and compared its accuracy with conventional prediction methods that uses the rate of estimated glomerular filtration rate (eGFR) decline. METHODS: Data of adult chronic kidney disease (CKD) patients who underwent hemodialysis at Oita University Hospital from April 2016 to March 2021 were extracted from electronic medical records (N = 135). A new machine learning predictor was compared with the established prediction method that uses the eGFR decline rate and the accuracy of the prediction models was determined using the coefficient of determination (R2). The data were preprocessed and split into training and validation datasets. We created multiple machine learning models using the training data and evaluated their accuracy using validation data. Furthermore, we predicted the time to RRT using a conventional prediction method that uses the eGFR decline rate for patients who had measured eGFR three or more times in two years and evaluated its accuracy. RESULTS: The least absolute shrinkage and selection operator regression model exhibited moderate accuracy with an R2 of 0.60. By contrast, the conventional prediction method was found to be extremely low with an R2 of -17.1. CONCLUSIONS: The significance of this study is that it shows that machine learning can predict time to RRT moderately well with continuous values from data at a single time point. This approach outperforms the conventional prediction method that uses eGFR time series data and presents new avenues for CKD treatment.

Two cases of idiopathic multicentric Castleman disease with nephrotic syndrome treated with tocilizumab.

We report two cases of idiopathic multicentric Castleman disease (iMCD) with nephrotic syndrome (NS) treated with tocilizumab. Case 1 was a 58-year-old man diagnosed with iMCD prior to the onset of NS. Renal biopsy revealed membranous nephropathy, which was considered to be secondary membranous nephropathy associated with iMCD. Case 2 was a 49-year-old woman diagnosed with iMCD prior to NS. Renal biopsy revealed renal amyloidosis positive for Congo red staining and amyloid A protein immunostaining. In both the cases, the proteinuria improved after the initiation of glucocorticoid and tocilizumab therapy. Tocilizumab may be a good therapeutic choice for iMCD with NS.

Concurrent chemo-radiotherapy for nasopharyngeal carcinoma.

OBJECTIVE: Because of the low incidence of nasopharyngeal carcinoma (NPC) in Japan, few studies have reported on the use of concurrent chemo-radiotherapy (CCRT) for treatment of this disease. The present study investigated the efficacy and toxicity of CCRT with Carboplatin (CBDCA) and UFT for NPC. PATIENTS AND METHODS: CCRT with injection of CBDCA and oral administration of UFT was given to 21 patients in our institution during 1996-2000. We compared the patients treated with CCRT with those treated with RT alone from 1976 to 1995 or those treated with RT after chemotherapy (Pre-RT chemotherapy) from 1985 to 1995. Overall survival rate (OS), progression free survival rate (PFS) and cause specific survival (CSS) rate were analyzed. RESULTS: The actual 2-year OS, PFS, and CSS of CCRT were 85.7, 66.7 and 90%, respectively. The rates were improved compared to RT alone (37.9, 31.0, and 56.6%) and Pre-RT chemotherapy (58.8, 41.2, and 64.7%). The most frequent and severe acute toxicities were leukopenia (42.9%) and mucositis with grade 3-4 (28.6%), both of which eventually resolved. CONCLUSION: CCRT with CBDCA and UFT is considered to be an effective, convenient, and tolerable treatment, which improves survival rates for NPC patients. Therefore, we recommend this CCRT method for the treatment of NPC.

Clinicopathological study of undifferentiated carcinoma of the parotid gland.

OBJECTIVE: Undifferentiated carcinoma of the salivary gland is a rare malignant tumor, and is difficult to distinguish from other poorly differentiated types of carcinoma or sarcoma. The present study investigated clinical and pathological characteristics for undifferentiated carcinoma of the parotid gland. PATIENTS AND METHODS: Forty-four patients with previously untreated carcinoma of the major salivary glands were treated at our institution between 1986 and 1999. Of these, five patients (two males, three females) were diagnosed with undifferentiated carcinoma of the parotid gland and treated. For histological reinvestigation, multislices of resected specimens were made and diagnosed. RESULTS: Although all patients were treated by radical surgery (more than subtotal parotidectomy), all died of distant metastasis, including two patients with locoregional relapse. Tumors in these two patients included a small portion of poorly differentiated epidermoid or mucoepidermoid carcinoma. The remaining three patients did not show any differentiated portions, and histological findings demonstrated heterogeneous patterns of lymphoepithelial carcinoma, small cell carcinoma and unclassified (a pattern of malignant hemangiopericytoma), respectively. CONCLUSIONS: Investigation using multislice sections is needed to diagnose undifferentiated carcinoma of the salivary glands. Regarding prognosis, carcinoma that is too poorly differentiated but including slightly-differentiated portions should be considered undifferentiated carcinoma. All patients died of distant metastasis despite radical surgery, suggesting that chemotherapy is needed to improve patient outcomes.