Machine learning to predict overall short-term mortality in cutaneous melanoma.

BACKGROUND: Cutaneous malignant melanoma (CMM) ranks among the ten most frequent malignancies, clinicopathological staging being of key importance to predict prognosis. Artificial intelligence (AI) has been recently applied to develop prognostically reliable staging systems for CMM. This study aims to provide a useful machine learning based tool to predict the overall CMM short-term survival. METHODS: CMM records as collected at the Veneto Cancer Registry (RTV) and at the Veneto regional health service were considered. A univariate Cox regression validated the strength and direction of each independent variable with overall mortality. A range of machine learning models (Logistic Regression classifier, Support-Vector Machine, Random Forest, Gradient Boosting, and k-Nearest Neighbors) and a Deep Neural Network were then trained to predict the 3-years mortality probability. Five-fold cross-validation and Grid Search were performed to test the best data preprocessing procedures, features selection, and to optimize models hyperparameters. A final evaluation was carried out on a separate test set in terms of balanced accuracy, precision, recall and F1 score. The best model was deployed as online tool. RESULTS: The univariate analysis confirmed the significant prognostic value of TNM staging. Adjunctive clinicopathological variables not included in the AJCC 8th melanoma staging system, i.e., sex, tumor site, histotype, growth phase, and age, were significantly linked to overall survival. Among the models, the Neural Network and the Random Forest models featured the best prognostic performance, achieving a balanced accuracy of 91% and 88%, respectively. According to the Gini importance score, age, T and M stages, mitotic count, and ulceration appeared to be the variables with the greatest impact on survival prediction. CONCLUSIONS: Using data from patients with CMM, we developed an AI algorithm with high staging reliability, on top of which a web tool was implemented ( unipd.link/melanomaprediction ). Being essentially based on routinely recorded clinicopathological variables, it can already be implemented with minimal effort and further tested in the current clinical practice, an essential phase for validating the model's accuracy beyond the original research context.

Isolated limb perfusion for the management limb threatening soft tissue sarcomas: The role of histological type on clinical outcomes.

BACKGROUND: Hyperthermic isolated limb perfusion (HILP) is an effective neoadjuvant treatment to avoid amputation in patients with locally advanced extremity soft tissue sarcomas (STS). We aimed to investigate whether STS histological type plays a role in predicting clinical outcomes. METHODS: This study reports a retrospective analysis of 125 patients with limb threatening STS (liposarcoma, n = 41; malignant peripheral nerve sheath tumor, n = 20; leiomyosarcoma, n = 20; miscellany, n = 44), who underwent HILP from 1990 through 2015 at our institution. The following endpoints were evaluated: tumor response (assessed by radiological imaging and histology), limb sparing rate, local progression-free survival (LPFS) and overall survival (OS). RESULTS: On average, overall (complete + partial) tumor response was significantly greater in patients affected with liposarcoma as compared to those with other histotypes (radiological response rate: 38/41, 92.7% vs 66/84, 78.6%, P-value: 0.048; mean histological necrosis: 83.6% vs 52.9%, P < 0.0001). Limb sparing rate was also higher among patients with liposarcoma as compared to other histotypes (39/41, 95.1% vs 62/84, 73.8%, P-value: 0.005). As regards survival, LPFS was similar across tumor types, whereas OS resulted significantly worse in patients with limb leiomyosarcoma (log-rank P-value: 0.009). CONCLUSIONS: HILP is a very effective treatment modality for limb threatening STS. In our series, liposarcoma appears to be the histological type most sensitive to HILP in terms of tumor response and thus limb sparing, which might help clinicians in the patient selection process.

Differences Regarding Branded HA in Italy, Part 2: Data from Clinical Studies on Knee, Hip, Shoulder, Ankle, Temporomandibular Joint, Vertebral Facets, and Carpometacarpal Joint.

OBJECTIVES: The aim of the current study is to collect scientific data on all branded hyaluronic acid (HA) products in Italy that are in use for intra-articular (IA) injection in osteoarthritis (OA) compared with that reported in the leaflet. METHODS: An extensive literature research was performed for all articles reporting data on the IA use of HA in OA. Selected studies were taken into consideration only if they are related to products based on HAs that are currently marketed in Italy with the specific joint indication for IA use in patients affected by OA. RESULTS: Sixty-two HA products are marketed in Italy: 30 products are indicated for the knee but only 8 were proved with some efficacy; 9 products were effective for the hip but only 6 had hip indication; 7 products proved to be effective for the shoulder but only 3 had the indication; 5 products proved effective for the ankle but only one had the indication; 6 products were effective for the temporomandibular joint but only 2 had the indication; only 2 proved effective for vertebral facet joints but only 1 had the indication; and 5 products proved effective for the carpometacarpal joint but only 2 had the indication. CONCLUSIONS: There are only a few products with some evidences, while the majority of products remain without proof. Clinicians and regulators should request postmarketing studies from pharmaceuticals to corroborate with that reported in the leaflet and to gather more data, allowing the clinicians to choose the adequate product for the patient.

Carbon fragmentation measurements and validation of the Geant4 nuclear reaction models for hadrontherapy.

Nuclear fragmentation measurements are necessary when using heavy-ion beams in hadrontherapy to predict the effects of the ion nuclear interactions within the human body. Moreover, they are also fundamental to validate and improve the Monte Carlo codes for their use in planning tumor treatments. Nowadays, a very limited set of carbon fragmentation cross sections are being measured, and in particular, to our knowledge, no double-differential fragmentation cross sections at intermediate energies are available in the literature. In this work, we have measured the double-differential cross sections and the angular distributions of the secondary fragments produced in the (12)C fragmentation at 62 A MeV on a thin carbon target. The experimental data have been used to benchmark the prediction capability of the Geant4 Monte Carlo code at intermediate energies, where it was never tested before. In particular, we have compared the experimental data with the predictions of two Geant4 nuclear reaction models: the Binary Light Ions Cascade and the Quantum Molecular Dynamic. From the comparison, it has been observed that the Binary Light Ions Cascade approximates the angular distributions of the fragment production cross sections better than the Quantum Molecular Dynamic model. However, the discrepancies observed between the experimental data and the Monte Carlo simulations lead to the conclusion that the prediction capability of both models needs to be improved at intermediate energies.

Use of disphosphonates in the treatment of osteoporosis in thyroidectomized patients on levothyroxin replacement therapy.

Bone mineral density at medio distal (MD-BMD) and ultra distal (UD-BMD) sites of non dominant radius were studied, using a DEXA Mineralometer (TURBOSCAN-NIM) on a series of 20 patients, over 60 years old, all thyroidectomized for thyroid carcinoma and treated with levothyroxin replacement therapy. T3 and T4 (RIA method), TSH (IRMA method) and two cardiac parameters as TPER (time peak ejection rate) and TPFR (time peak filling rate) by angiocardioscintigraphy were also determined. Results showed that 19 patients considering MD-BMD values and ten considering UD-BMD values were at risk of fractures. Circulating T3 levels were within normal range in 17 patients, elevated in two cases and reduced in one case. Circulating T4 levels were within normal range in 15 patients, increased in four cases and reduced in one case. Circulating TSH levels were within normal range in 17 patients, reduced in two cases and elevated in one case. TPER were reduced in 18 patients and normal in two cases. TPFR were reduced in six patients and normal in fourteen cases. The authors administered alendronate (10 mg/day) which seems to prevent bone loss, especially at the level of the trabecular bone. Bone loss reverted at 6 and 12 months, confirming that disphosphonates slow down this phenomenon in a situation of increased bone turnover (e.g. patients on long-term L-T4 therapy).

The use of bisphosphonates in the treatment of osteoporosis.

Three therapeutic protocols using 3 different bisphosphonate molecules have been evaluated in three groups of women in physiological menopause, whose bone mineral density (BMD) at the ultradistal (UD) or mediodistal (MD) radius showed risk of fracture. Treatment with alendronate (5 mg/day, for 12 months) increases BMD (38 patients, mean age 63 +/- 6 years) by about 5%; treatment with ethidronate (400 mg/day form days, then 1.250 mg/day calcium carbonate for 76 days), repeated 4-times (31 patients, mean age 66 +/- 6 years) did not cause any significant rise in BMD of either UD or MD radius; treatment with clodronate (10 mg/day i.m. for 20 days, a 10-day-interval, followed by 400 mg/day for 60 days) resulted in an insignificant increase (about 4%) in the BMDUD value (8 patients, mean age 62 +/- 5 years). No vertebral fractures nor deformations were observed in any of the 3 groups during treatments. All groups showed an optimal tolerance to the drugs.

Senile diabetes and bone mineral density.

Bone mineral density was determined in a series of 67 elderly diabetics (38 males and 29 females) and 40 non-diabetic elderly subjects (20 males and 20 females) at the third medial and tenth ultradistal of the non-dominating radius using an X-ray densitometer (DEXA). Bone metabolism markers (Ct, PTH, HOP, UCA, AP, Vit-25-OH-D, BGP) were also measured. Our results indicate that there is no significant difference in values of BMD and the bone metabolism markers studied between diabetic and non-diabetic elderly subjects. We believe that senile diabetes is not a risk factor of onset and maintenance of senile osteoporosis.

Ipriflavone-treatment of senile osteoporosis: results of a multicenter, double-blind clinical trial of 2 years.

Eighty-four out of 100 enrolled female patients affected by osteoporosis completed a double-blind, randomized trial, using ipriflavone (IP) in a dose of 3 x 200 mg/day (41 patients) or placebo (P)(43 patients). All patients received 1 g/day oral calcium supplementation. Inclusion criteria were: age over 65 years; at least one vertebral fracture in the past; bone mineral density measured at the distal tenth of the radius lower than the normal average -2 x S.D. The trial period was between June 1990 and November 1993. Patients of the IP group showed a significant increase in bone mineral density (P < 0.05) compared with the starting values during the whole study period. Pain decreased rapidly, intake of analgesics dropped, and often significant decreases in calciuria, hydroxyprolinuria, alkaline phospatase, osteocalcin and parathormone values were observed. Only two new fractures occurred during the trial in the IP group. Patients of the P group at the end of the study displayed decreased bone mineral densities (P < 0.05), increased pain, greater consumption of analgesics, and often significant increases in the bone metabolism parameters listed for the IP group. In the