ML Models Built Using Clinical Parameters and Radiomic Features Extracted from 18F-Choline PET/CT for the Prediction of Biochemical Recurrence after Metastasis-Directed Therapy in Patients with Oligometastatic Prostate Cancer.

Oligometastatic patients at [18F]F-Fluorocholine (18F-choline) PET/CT may be treated with metastasis-directed therapy (MDT). The aim of this study was to combine radiomic parameters extracted from 18F-choline PET/CT and clinical data to build machine learning (ML) models able to predict MDT efficacy. METHODS: Oligorecurrent patients (≤5 lesions) at 18F-choline PET/CT and treated with MDT were collected. A per-patient and per-lesion analysis was performed, using 2-year biochemical recurrence (BCR) after MDT as the standard of reference. Clinical parameters and radiomic features (RFts) extracted from 18F-choline PET/CT were used for training five ML Models for both CT and PET images. The performance metrics were calculated (i.e., Area Under the Curve-AUC; Classification Accuracy-CA). RESULTS: A total of 46 metastases were selected and segmented in 29 patients. BCR after MDT occurred in 20 (69%) patients after 2 years of follow-up. In total, 73 and 33 robust RFTs were selected from CT and PET datasets, respectively. PET ML Models showed better performances than CT Models for discriminating BCR after MDT, with Stochastic Gradient Descent (SGD) being the best model (AUC = 0.95; CA = 0.90). CONCLUSION: ML Models built using clinical parameters and CT and PET RFts extracted via 18F-choline PET/CT can accurately predict BCR after MDT in oligorecurrent PCa patients. If validated externally, ML Models could improve the selection of oligorecurrent PCa patients for treatment with MDT.

Anti-predator behavioral responses of Italian agile frog tadpoles (Rana latastei) exposed to microplastics.

Microplastics (MPs) are nowadays abundant, persistent, and ubiquitous in the environment, representing a new threat for terrestrial, marine, and freshwater ecosystems. Although anuran populations and species are globally declining, the effect of MP exposure on this taxon has been poorly investigated. With the aim of assessing the effects of microplastic exposure on the defensive responses of Italian agile frog (Rana latastei) tadpoles, we exposed them to three different concentrations (1, 7, and 50 mg L-1) of a mixture of plastic polymers (HPDE, PVC, PS, and PES) for 2 weeks. Then, we measured the total distance covered by individual tadpoles before and after exposure to tadpole-fed dragonfly larvae (Aeshna cyanea) cues. As expected, predation risk sharply lowered the total distance travelled by tadpoles; however, MP concentration did not affect their defensive performances. We also collected data on tadpole development, activity, and mortality. In contrast with previous experiments, neither tadpole growth nor mortality varied with MP concentration. Our results indicate that the intensity of MP effects on growth and development may depend on tadpole size, with large tadpoles being less susceptible to the negative effects of MP exposure.

Differential effects of microplastic exposure on anuran tadpoles: A still underrated threat to amphibian conservation?

Microplastics (MPs) have been reported to threaten a wide variety of terrestrial, marine, and freshwater organisms. However, knowledge about the effects of MPs on anuran amphibians, one of the most threatened taxa worldwide, is still limited. To assess the effects of MPs on the growth and survival of the Italian agile frog (Rana latastei) and green toad (Bufotes balearicus), we exposed tadpoles to three different concentrations (1, 7, and 50 mg L-1) of an environmental relevant mixture of microplastics (HPDE, PVC, PS and PES), recording data on their activity level, weight and mortality rates. While the effects of MPs on green toad tadpoles were negligible, Italian agile frog tadpoles were severely affected both in terms of growth and activity level, with high mortality rates even at the lowest MP density (1 mg L-1). Our results suggest that MP contamination of freshwater habitats may contribute to the ongoing decline of anuran amphibians.

Contextual behavioural plasticity in Italian agile frog (Rana latastei) tadpoles exposed to native and alien predator cues.

Predation is a strong driver for the evolution of prey behaviour. To properly assess the actual risk of predation, anuran tadpoles mostly rely on water-borne chemical cues, and their ability to evaluate environmental information is even more crucial when potential predators consist of unknown alien species. Behavioural plasticity - that is, the capacity to express changes in behaviour in response to different environmental stimuli - is crucial to cope with predation risk. We explored the defensive behaviour of Italian agile frog (Rana latastei) tadpoles when exposed to the chemical cues of two predator species, one native (dragonfly larvae) and one alien (red swamp crayfish). Firstly, we observed whether a plastic life history trait (i.e. hatching time) might be affected by native predatory cues. Secondly, we recorded a suite of behavioural responses (activity level, lateralization and sinuosity) to each cue. For assessing lateralization and sinuosity, we developed a C++ code for the automatic analysis of digitally recorded tadpole tracks. Hatching time seemed not to be affected by the potential risk of predation, while both predator species and diet affected tadpoles' defensive behaviour. Tadpoles responded to a predator threat by two main defensive strategies: freezing and 'zig-zagging'. While the first behaviour had previously been reported, the analysis of individual trajectories indicated that tadpoles can also increase path complexity, probably to prevent predators from anticipating their location. We also recorded a decrease in lateralization intensity, which suggests that under predation risk, tadpoles tend to scrutinize the surrounding environment equally on both sides.

A massive update of non-indigenous species records in Mediterranean marinas.

The Mediterranean Sea is home to over 2/3 of the world's charter boat traffic and hosts an estimated 1.5 million recreational boats. Studies elsewhere have demonstrated marinas as important hubs for the stepping-stone transfer of non-indigenous species (NIS), but these unique anthropogenic, and typically artificial habitats have largely gone overlooked in the Mediterranean as sources of NIS hot-spots. From April 2015 to November 2016, 34 marinas were sampled across the following Mediterranean countries: Spain, France, Italy, Malta, Greece, Turkey and Cyprus to investigate the NIS presence and richness in the specialized hard substrate material of these marina habitats. All macroinvertebrate taxa were collected and identified. Additionally, fouling samples were collected from approximately 600 boat-hulls from 25 of these marinas to determine if boats host diverse NIS not present in the marina. Here, we present data revealing that Mediterranean marinas indeed act as major hubs for the transfer of marine NIS, and we also provide evidence that recreational boats act as effective vectors of spread. From this wide-ranging geographical study, we report here numerous new NIS records at the basin, subregional, country and locality level. At the basin level, we report three NIS new to the Mediterranean Sea (Achelia sawayai sensu lato, Aorides longimerus, Cymodoce aff. fuscina), and the re-appearance of two NIS previously known but currently considered extinct in the Mediterranean (Bemlos leptocheirus, Saccostrea glomerata). We also compellingly update the distributions of many NIS in the Mediterranean Sea showing some recent spreading; we provide details for 11 new subregional records for NIS (Watersipora arcuata, Hydroides brachyacantha sensu lato and Saccostrea glomerata now present in the Western Mediterranean; Symplegma brakenhielmi, Stenothoe georgiana, Spirobranchus tertaceros sensu lato, Dendostrea folium sensu lato and Parasmittina egyptiaca now present in the Central Mediterranean, and W. arcuata, Bemlos leptocheirus and Dyspanopeus sayi in the Eastern Mediterranean). We also report 51 new NIS country records from recreational marinas: 12 for Malta, 10 for Cyprus, nine for Greece, six for Spain and France, five for Turkey and three for Italy, representing 32 species. Finally, we report 20 new NIS records (representing 17 species) found on recreational boat-hulls (mobile habitats), not yet found in the same marina, or in most cases, even the country. For each new NIS record, their native origin and global and Mediterranean distributions are provided, along with details of the new record. Additionally, taxonomic characters used for identification and photos of the specimens are also provided. These new NIS records should now be added to the relevant NIS databases compiled by several entities. Records of uncertain identity are also discussed, to assess the probability of valid non-indigenous status.

Correction: A massive update of non-indigenous species records in Mediterranean marinas.

[This corrects the article DOI: 10.7717/peerj.3954.].