Deep Learning-Based Objective and Reproducible Osteosarcoma Chemotherapy Response Assessment and Outcome Prediction.

Osteosarcoma is the most common primary bone cancer, whose standard treatment includes pre-operative chemotherapy followed by resection. Chemotherapy response is used for prognosis and management of patients. Necrosis is routinely assessed after chemotherapy from histology slides on resection specimens, where necrosis ratio is defined as the ratio of necrotic tumor/overall tumor. Patients with necrosis ratio ≥90% are known to have a better outcome. Manual microscopic review of necrosis ratio from multiple glass slides is semiquantitative and can have intraobserver and interobserver variability. In this study, an objective and reproducible deep learning-based approach was proposed to estimate necrosis ratio with outcome prediction from scanned hematoxylin and eosin whole slide images (WSIs). To conduct the study, 103 osteosarcoma cases with 3134 WSIs were collected. Deep Multi-Magnification Network was trained to segment multiple tissue subtypes, including viable tumor and necrotic tumor at a pixel level and to calculate case-level necrosis ratio from multiple WSIs. Necrosis ratio estimated by the segmentation model highly correlates with necrosis ratio from pathology reports manually assessed by experts. Furthermore, patients were successfully stratified to predict overall survival with P = 2.4 × 10-6 and progression-free survival with P = 0.016. This study indicates that deep learning can support pathologists as an objective tool to analyze osteosarcoma from histology for assessing treatment response and predicting patient outcome.

Burden of Illness Associated with Tenosynovial Giant Cell Tumors.

PURPOSE: Little is known about the burden of illness in patients with tenosynovial giant cell tumors (TGCT), which are rare, typically benign, lesions of the synovial tissue including giant cell tumor of the tendon sheath (GCT-TS) and pigmented villonodular synovitis (PVNS). The objective of this study was to describe health care resource use and costs for patients with GCT-TS and PVNS, which are rare and typically benign TGCT. METHODS: A retrospective cohort study design was used to analyze administrative claims for adult commercial and Medicare Advantage health plan enrollees with evidence of GCT-TS and PVNS from January 1, 2006 through March 31, 2015. Participants were continuously enrolled for 12 months before (pre-index period) and 12 months after (post-index period) the date of the first tenosynovial giant cell tumor (TGCT) claim (index date). Preindex and postindex measures were compared using the McNemar test and Wilcoxon signed-rank test. Results were stratified by TGCT type. FINDINGS: The study identified 4664 patients with TGCT, 284 with GCT-TS, and 4380 with PVNS. Mean age (GCT-TS group: 50 years; PVNS group: 51 years) and sex distributions (GCT-TS group: 60.2% female; PVNS group: 59.5% female) were similar for each group. Most patients with GCT-TS (78.2%) had at least one postindex surgery, compared with 38.7% of patients with PVNS. Mean total health care costs increased from $8943 in the preindex period to $14,880 in the postindex period (P < 0.001) for GCT-TS and from $13,221 in the preindex period to $17,728 in the postindex period (P < 0.001) for PVNS. Preindex to postindex ambulatory costs increased nearly 120% for patients with GCT-TS ($4340 to $9570, P < 0.001) and 50% for patients with PVNS ($6782 to $10,278, P < 0.001), and physical therapy use increased significantly during the same period (GCT-TS: 18% to 40%, P < 0.001; PVNS: 38% to 60%, P < 0.001). IMPLICATIONS: Costs increased substantially 1 year after the first TGCT claim, with more than half the costs covering ambulatory care. These results suggest a high health care burden once TGCT is identified.

How do we estimate survival? External validation of a tool for survival estimation in patients with metastatic bone disease-decision analysis and comparison of three international patient populations.

BACKGROUND: We recently developed a clinical decision support tool, capable of estimating the likelihood of survival at 3 and 12 months following surgery for patients with operable skeletal metastases. After making it publicly available on www.PATHFx.org , we attempted to externally validate it using independent, international data. METHODS: We collected data from patients treated at 13 Italian orthopaedic oncology referral centers between 2010 and 2013, then applied to PATHFx, which generated a probability of survival at three and 12-months for each patient. We assessed accuracy using the area under the receiver-operating characteristic curve (AUC), clinical utility using Decision Curve Analysis (DCA), and compared the Italian patient data to the training set (United States) and first external validation set (Scandinavia). RESULTS: The Italian dataset contained 287 records with at least 12 months follow-up information. The AUCs for the three-month and 12-month estimates was 0.80 and 0.77, respectively. There were missing data, including the surgeon's estimate of survival that was missing in the majority of records. Physiologically, Italian patients were similar to patients in the training and first validation sets. However notable differences were observed in the proportion of those surviving three and 12-months, suggesting differences in referral patterns and perhaps indications for surgery. CONCLUSIONS: PATHFx was successfully validated in an Italian dataset containing missing data. This study demonstrates its broad applicability to European patients, even in centers with differing treatment philosophies from those previously studied.

Making a case for the socioeconomic determinacy of survival in osteosarcoma.

BACKGROUND: The literature on osteosarcoma survival generally focuses on tumor and treatment variables, although it is unclear whether and how ethnic and socioeconomic factors might influence survival. QUESTIONS/PURPOSES: We therefore investigated the relative contribution of socioeconomic influences together with more traditional tumor-specific factors on osteosarcoma survival. METHODS: We performed survival analyses on two national databases in two countries. Using multivariable analyses, we compared these with corresponding institution-specific survival to determine if socioeconomic factors might impact osteosarcoma survival. RESULTS: East Asian descent, state-specific treatment, female sex, treatment in the 1990s, low-grade disease, intracompartmental disease, small size, wide resections as opposed to forequarter or hindquarter amputations, and single primaries were good prognostic factors. Survival was better in the more affluent states. Males were affected at an older age than females. Blacks tended to have larger tumors, although their overall survival was similar to whites. East Asians were more likely to be treated in the 1990s with wide resections for smaller tumors and were located around states associated with good treatment. East Asians in Singapore and the United States had the same survival. Survival in East Asians in Singapore was similar to that of other races. The provision of health care for osteosarcoma varies greatly across the United States but is uniform in the socialized medical system in Singapore. Hence, the observed differences in the United States were likely the result of socioeconomic factors. CONCLUSIONS: Our analysis suggests ethnic and economic bias may influence survival in osteosarcoma and should receive greater attention in the collective literature on survival analyses. LEVEL OF EVIDENCE: Level II, prognostic study. See the Guidelines for Authors for a complete description of levels of evidence.

Histopathologic and Radiologic Assessment of Chemotherapeutic Response in Ewing's Sarcoma: A Review.

Ewing's sarcoma is a highly malignant tumor that metastasizes rapidly and is thus associated with a low survival rate. The intensification of chemotherapy has been shown to improve the overall survival of patients with Ewing's sarcoma. However, intensified chemotherapy can lead to increased toxicity or even the development of secondary malignancies. The stratification of patients with Ewing's sarcoma into "good" and "poor" responders may help guide the administration of progressively more intensified chemotherapy. Thus, an accurate assessment of the chemotherapeutic response, as well as the extent of chemotherapy-induced tumor necrosis, is critical for avoiding potential treatment-related complications in these patients. This paper reviews the methods currently used to evaluate chemotherapeutic response in Ewing's sarcoma, focusing specifically on histopathologic and imaging analyses, and discusses novel therapies and imaging methods that may help improve the overall survival of these patients.

Direct-push geochemical profiling for assessment of inorganic chemical heterogeneity in aquifers.

Discrete-depth sampling of inorganic groundwater chemistry is essential for a variety of site characterization activities. Although the mobility and rapid sampling capabilities of direct-push techniques have led to their widespread use for evaluating the distribution of organic contaminants, complementary methods for the characterization of spatial variations in geochemical conditions have not been developed. In this study, a direct-push-based approach for high-resolution inorganic chemical profiling was developed at a site where sharp chemical contrasts and iron-reducing conditions had previously been observed. Existing multilevel samplers (MLSs) that span a fining-upward alluvial sequence were used for comparison with the direct-push profiling. Chemical profiles obtained with a conventional direct-push exposed-screen sampler differed from those obtained with an adjacent MLS because of sampler reactivity and mixing with water from previous sampling levels. The sampler was modified by replacing steel sampling components with stainless-steel and heat-treated parts, and adding an adapter that prevents mixing. Profiles obtained with the modified approach were in excellent agreement with those obtained from an adjacent MLS for all constituents and parameters monitored (Cl, NO(3), Fe, Mn, DO, ORP, specific conductance and pH). Interpretations of site redox conditions based on field-measured parameters were supported by laboratory analysis of dissolved Fe. The discrete-depth capability of this approach allows inorganic chemical variations to be described at a level of detail that has rarely been possible. When combined with the mobility afforded by direct-push rigs and on-site methods of chemical analysis, the new approach is well suited for a variety of interactive site-characterization endeavors.