Validation of PATHFx 2.0: An open-source tool for estimating survival in patients undergoing pathologic fracture fixation

Treatment decisions in patients with metastatic bone disease rely on accurate survival estimation. We developed the original PATHFx models using expensive, proprietary software and now seek to provide a more cost-effective solution. Using open-source machine learning software to create PATHFx version 2.0, we asked whether PATHFx 2.0 could be created using open-source methods and externally validated in two unique patient populations. The training set of a well-characterized, database records of 189 patients and the bnlearn package within R Version 3.5.1 (R Foundation for Statistical Computing), was used to establish a series of Bayesian belief network models designed to predict survival at 1, 3, 6, 12, 18, and 24 months. Each was externally validated in both a Scandinavian (n = 815 patients) and a Japanese (n = 261 patients) data set. Brier scores and receiver operating characteristic curves to assessed discriminatory ability. Decision curve analysis (DCA) evaluated whether models should be used clinically. DCA showed that the model should be used clinically at all time points in the Scandinavian data set. For the 1-month time point, DCA of the Japanese data set suggested to expect better outcomes assuming all patients will survive greater than 1 month. Brier scores for each curve demonstrate that the models are accurate at each time point. Statement of Clinical Significance: we successfully transitioned to PATHFx 2.0 using open-source software and externally validated it in two unique patient populations, which can be used as a cost-effective option to guide surgical decisions in patients with metastatic bone disease.     

Long-term effects of adjuvant tamoxifen and/or radiotherapy. The South Sweden Breast Cancer Trial.

In a multicenter trial of adjuvant therapy in stage II breast cancer, 719 postmenopausal patients were randomized to one of three treatment regimens: radiotherapy only or in combination with adjuvant tamoxifen for one year, or adjuvant tamoxifen without radiotherapy. At twelve years of follow-up (median 9 years), no statistically significant differences in survival or recurrence-free survival were observed. However, the rate of loco-regional recurrency was lower among patients treated with both radiotherapy and tamoxifen. The rate of bilateral breast cancer was reduced in tamoxifen-treated patients whereas the rate of new primary malignancies other than breast cancer was somewhat higher in tamoxifen-treated patients. Adjuvant therapy in breast cancer may influence not only breast cancer recurrences and mortality but also later disease patterns and cause-specific mortality.     

Long-term effects on pulmonary function of mantle radiotherapy in patients with Hodgkin's disease.

Twenty-five patients (21-45 years old) treated for Hodgkin's disease with mantle radiotherapy but no chemotherapy underwent chest radiography and pulmonary testing with spirometry, pulmonary mechanics and exercise test combined with arterial blood gas analysis, lung scintigraphy, assessment of pulmonary artery pressure with Doppler cardiography and vector ECG 10-20 years after treatment. The doses to mediastinum ranged from 35-43 (mean 40) Gy given in 26 fractions with the split-course technique. Radiographic signs of slight to moderate pulmonary fibrosis were seen in 18 patients. Minor restrictive ventilatory defects were found with decreased VC, TLC and lung compliance and increased maximal elastic recoil. Little evidence of airflow obstruction was found. Exercise capacity was decreased in three individuals but the mean value for the study group as a whole was normal. Arterial PO2 at maximum exercise was reduced but no patient had diminished hemoglobin saturation. Lung scintigraphy showed defects in 21 patients, mostly consisting of slight abnormalities at the lung periphery and apices. The perfusion seemed to be more affected than the ventilation, suggesting primary vascular lesions. Twelve patients showed signs of right ventricular hypertrophy in vector ECG and four of these had systolic pulmonary artery pressure greater than or equal to 30 mm Hg. The observed abnormalities were mostly of a minor degree and few clinically significant long-term effects of mantle radiotherapy on pulmonary function were observed.     

Bb2Bb3 regulation of murine Lyme arthritis is distinct from Ncf1 and independent of the phagocyte nicotinamide adenine dinucleotide phosphate oxidase

Several quantitative trait loci regulating murine Lyme arthritis severity have been mapped, including a highly significant linkage found on chromosome 5, termed Bb2Bb3. Within this region, the Ncf1 gene of the phagocyte nicotinamide adenine dinucleotide phosphate (NADPH) oxidase has recently been identified as a major regulator of arthritis severity in rodent models of rheumatoid arthritis, an effect attributed to protective properties of reactive oxygen species. To assess the role of Ncf1 in Lyme arthritis, we introgressed Bb2Bb3 from severely arthritic C3H/He mice onto mildly arthritic C57BL/6 mice. This increased Lyme arthritis severity, whereas the reciprocal transfer conferred protection from disease. A single nucleotide polymorphism was identified in the Ncf1 gene that did not influence the protein sequence or expression of Ncf1. Although polymorphonuclear leukocytes from C57BL/6 mice generated a greater oxidative burst than polymorphonuclear leukocytes from C3H/He mice, studies with the Bb2Bb3 congenic mice demonstrated this difference was not linked to Ncf1 alleles. Furthermore, Lyme arthritis severity was not altered in mice lacking either the Ncf1 or Gp91phox subunits of the NADPH oxidase complex. Together, these results argue that Ncf1 is not a candidate gene for regulation of Lyme arthritis and reveal Lyme arthritis to be independent of NADPH oxidase activity, distinguishing it from other models of rheumatoid arthritis.     

IL-4-deficient mice develop less acute but more chronic relapsing collagen-induced arthritis

Rheumatoid arthritis as well as collagen-induced arthritis (CIA) is thought to involve T cell autoimmunity of the Th1 type and the Th2 cytokine IL-4 has been proposed to play a suppressive role. To exclude a possible skewing role of the mycobacteria used in the complete Freund's adjuvant (CFA) we induced CIA with type II collagen (CII) in incomplete Freund's adjuvant (IFA). Our results show that IL-4 deficiency leads to a lesser susceptibility to arthritis and lower B and T cell responses if induced with CII/IFA but not if induced with CII/CFA. In addition, IL-4-deficient mice were less susceptible to arthritis induced with monoclonal anti-CII antibodies. However, mice immunized with CII/IFA later developed a chronic relapsing disease, which was promoted by IL-4 deficiency. We conclude that IL-4 plays different roles depending on the type of adjuvant used and the phase (acute or chronic) of the clinical disease.     

Assessing the consequences of quarantines during a pandemic

The European Journal of Health Economics - This paper analyzes the epidemiological and economic effects of quarantines. We use a basic epidemiological model, a SEIR-model, that is calibrated to...     

The major T cell epitope on type II collagen is glycosylated in normal cartilage but modified by arthritis in both rats and humans

Type II collagen (CII) is a target for autoreactive T cells in both rheumatoid arthritis and the murine model collagen-induced arthritis. The determinant core of CII has been identified as CII260-270, and the alteration of this T cell epitope by posttranslational modifications is known to be critical for development of arthritis in mice. Using CII-specific T cell hybridomas we have now shown that the immunodominant T cell epitope in the normal (healthy) human and rat joint cartilage is O-glycosylated at the critical T cell receptor recognition position 264 with a mono- or di-saccharide attached to a hydroxylysine. In contrast, in the arthritic human and rat joint cartilage there are both glycosylated and non-glycosylated CII forms. Glycosylated CII from normal cartilage could not be recognized by T cells reactive to peptides having only lysine or hydroxylysine at position 264, showing that antigen-presenting cells could not degrade the O-linked carbohydrate. Thus, the variable forms of the glycosylated epitope are determined by the structures present in cartilage, and these vary during the disease course. We conclude that the chondrocyte determines the structures presented to the immune system and that these structures are different in normal versus arthritic states.