Deep learning workflow to support in-flight processing of digital aerial imagery for wildlife population surveys.

Deep learning shows promise for automating detection and classification of wildlife from digital aerial imagery to support cost-efficient remote sensing solutions for wildlife population monitoring. To support in-flight orthorectification and machine learning processing to detect and classify wildlife from imagery in near real-time, we evaluated deep learning methods that address hardware limitations and the need for processing efficiencies to support the envisioned in-flight workflow. We developed an annotated dataset for a suite of marine birds from high-resolution digital aerial imagery collected over open water environments to train the models. The proposed 3-stage workflow for automated, in-flight data processing includes: 1) image filtering based on the probability of any bird occurrence, 2) bird instance detection, and 3) bird instance classification. For image filtering, we compared the performance of a binary classifier with Mask Region-based Convolutional Neural Network (Mask R-CNN) as a means of sub-setting large volumes of imagery based on the probability of at least one bird occurrence in an image. On both the validation and test datasets, the binary classifier achieved higher performance than Mask R-CNN for predicting bird occurrence at the image-level. We recommend the binary classifier over Mask R-CNN for workflow first-stage filtering. For bird instance detection, we leveraged Mask R-CNN as our detection framework and proposed an iterative refinement method to bootstrap our predicted detections from loose ground-truth annotations. We also discuss future work to address the taxonomic classification phase of the envisioned workflow.

Venetoclax in Previously Treated Waldenström Macroglobulinemia.

PURPOSE: BCL2 is overexpressed and confers prosurvival signaling in malignant lymphoplasmacytic cells in Waldenström macroglobulinemia (WM). Venetoclax is a potent BCL2 antagonist and triggers in vitro apoptosis of WM cells. The activity of venetoclax in WM remains to be clarified. PATIENTS AND METHODS: We performed a multicenter, prospective phase II study of venetoclax in patients with previously treated WM (NCT02677324). Venetoclax was dose-escalated from 200 mg to a maximum dose of 800 mg daily for up to 2 years. RESULTS: Thirty-two patients were evaluable, including 16 previously exposed to Bruton tyrosine kinase inhibitors (BTKis). All patients were MYD88 L265P-mutated, and 17 carried CXCR4 mutations. The median time to minor and major responses was 1.9 and 5.1 months, respectively. Previous exposure to BTKis was associated with a longer time to response (4.5 v 1.4 months; P < .001). The overall, major, and very good partial response rates were 84%, 81%, and 19%, respectively. The major response rate was lower in those with refractory versus relapsed disease (50% v 95%; P = .007). The median follow-up time was 33 months, and the median progression-free survival was 30 months. CXCR4 mutations did not affect treatment response or progression-free survival. The only recurring grade ≥ 3 treatment-related adverse event was neutropenia (n = 14; 45%), including one episode of febrile neutropenia. Laboratory tumor lysis without clinical sequelae occurred in one patient. No deaths have occurred. CONCLUSION: Venetoclax is safe and highly active in patients with previously treated WM, including those who previously received BTKis. CXCR4 mutation status did not affect treatment response.

Response and survival predictors in a cohort of 319 patients with Waldenström macroglobulinemia treated with ibrutinib monotherapy.

Bruton tyrosine kinase (BTK) inhibitors are the only FDA-approved treatments for Waldenström macroglobulinemia (WM). Factors prognostic of survival and predictive of response to BTK inhibitors remained to be clarified. We evaluated 319 patients with WM to identify predictive and prognostic factors on ibrutinib monotherapy. Logistic and Cox proportional-hazard regression models were fitted for response and survival. Multiple imputation analyses were used to address bias associated with missing data. Major (partial response or better) and deep responses (very good partial response or better) were attained in 78% and 28% of patients. CXCR4 mutations were associated with lower odds of major (odds ratio [OR], 0.2; 95% confidence interval [CI], 0.1-0.5; P < .001) and deep response (OR, 0.3; 95% CI, 0.2-0.6; P = .001). CXCR4 mutations (hazard ratio [HR], 2.0; 95% CI, 1.2-3.4; P = .01) and platelet count 100 K/uL or less (HR, 2.5; 95% CI, 1.3-4.9; P = .007) were associated with worse progression-free survival (PFS). We proposed a scoring system using these 2 factors. The median PFS for patients with 0, 1, and 2 risk factors were not reached, 5 years and 3 years (P < .001). Patients with 2 risk factors had HR 2.2 (95% CI, 1.3-3.8; P = .004) compared with 1 factor, and patients with 1 factor had HR 2.3 (95% CI, 1.1-5.1; P = .03) compared with 0 factors. Age ≥65 years was the only factor associated with overall survival (HR, 3.2; 95% CI, 1.4-7.0; P = .005). Multiple imputation analyses did not alter our results. Our study confirms the predictive and prognostic value of CXCR4 mutations in patients with WM treated with ibrutinib monotherapy.

Phase 1 study of ibrutinib and the CXCR4 antagonist ulocuplumab in CXCR4-mutated Waldenström macroglobulinemia.

MYD88 and CXCR4 mutations are common in Waldenström macroglobulinemia (WM). Mutated CXCR4 (CXCR4Mut) impacts BTK-inhibitor response. We conducted a phase 1 trial of the CXCR4-antagonist ulocuplumab with ibrutinib in this first-ever study to target CXCR4Mut in WM. Ibrutinib was initiated at 420 mg/d with cycle 1 and continued until intolerance or progression; ulocuplumab was given cycles 1 to 6, with a 3 + 3 dose-escalation design. Each cycle was 4 weeks. Thirteen symptomatic patients, of whom 9 were treatment-naive patients were enrolled. Twelve were evaluable for response. At best response, their median serum immunoglobulin M declined from 5574 to 1114 mg/dL; bone marrow disease decreased from 65% to 10%, and hemoglobin increased from 10.1 to 14.2 g/dL (P < .001). The major and VGPR response rates were 100% and 33%, respectively, with VGPRs observed at lower ulocuplumab dose cohorts. Median times to minor and major responses were 0.9 and 1.2 months, respectively. With a median follow-up of 22.4 months, the estimated 2-year progression-free survival was 90%. The most frequent recurring grade ≥2 adverse events included reversible thrombocytopenia, rash, and skin infections. Ulocuplumab dose-escalation did not impact adverse events. The study demonstrates the feasibility of combining a CXCR4-antagonist with ibrutinib and provides support for the development of CXCR4-antagonists for CXCR4Mut WM. This trial was registered at www.clinicaltrials.gov as #NCT03225716.

Saltwater spray as an agent of natural selection: no evidence of local adaptation within a coastal population of Triplasis purpurea (Poaceae).

An ability to tolerate airborne saltwater spray is critical for plant populations in coastal environments. The opportunity for continued microevolution for improved salt tolerance can exist if there is variation in the response of genetic families to saltwater spray. Our objective was to determine whether or not there was differentiation among subpopulations near (15 m) and far (80 m) from shore and among families within subpopulations in relation to the effects of salt spray on life history traits in a population of the dunegrass Triplasis purpurea. In this annual, most seeds are matured in cleistogamous spikelets on axillary, leaf-sheath enclosed panicles and show poor dispersal capacity. Plants were reared in the greenhouse from seeds of 13 and 11 families from the near and far subpopulations, respectively. Fifty percent of plants in a family were subjected to 6 seawater sprays/wk, resulting in weekly salt deposition of 213 µg/cm(2); the others were sprayed with distilled water. Data were recorded on life span, tiller numbers, root and shoot dry mass, and seed production. There was no effect of subpopulation on any measured trait and, hence, no evidence for local adaptation to salt spray. Final tiller numbers, but not dry mass or seed production, were reduced by salt spray. However, for most traits there were significant family (within subpopulation) effects, indicating genetic substructuring. Life span and mean seed mass showed a significant family by treatment interaction, indicating genetic variation in phenotypic responses to salt spray. Life span and mean seed mass were reduced by salt spray in some, but not all, families. Path analysis revealed that an increase in life span or tiller number indirectly increased seed production via direct effects on vegetative mass. For this relatively salt-tolerant T. purpurea population on the south shore of Staten Island, New York, USA, salt sprays may not be a significant agent of natural selection. However, there are pronounced phenotypic differences among inbred family groups and opportunity for genetic substructuring within these subpopulations. Variable effects of salt spray among families could result in microevolutionary changes in life span and mean seed mass, both of which impact annual fitness in this dunegrass.