Using Gated Recurrent Unit Networks for the Prediction of Hemodynamic and Pulmonary Decompensation.

This paper presents a new medical severity scoring system, used to assess the risk of hemodynamic and pulmonary decompensation for patients being treated in intensive care units. The score presented here includes drug circulatory support and ventilation mode data for the evaluation of the patient's biosignals and laboratory values. It is shown that Gated Recurrent Unit-based neural networks are able to predict the maximal severity class within a 24 hour prediction time-frame (hemodynamic: 0.85 AUROC / pulmonary: 0.9 AUROC), and can estimate the underlying decompensation score for prediction times of up to 24 hours with mean errors of 6.3% of the maximal possible pulmonary, and 9.6% of the hemodynamic score. These results are based on 60h observation period. Clinical Relevance- Hemodynamic and pulmonary decom-pensation are life threatening dynamic events that can lead to death of patients. Early detection of these incidents is essential in order to intervene therapeutically and to improve survival chances. In everyday intensive care physicians are confronted with a vast number of laboratory values and vital parameters. There is a risk that early stages of hemodynamic and pulmonary decompensation are misjudged. The implementation of robust warning systems could support physicians in detecting these critical events and initiate therapeutical intervention in time which would achieve significant reduction of patient mortality.

Survival following disease recurrence of primary localized alveolar rhabdomyosarcoma.

BACKGROUND: Recurrences in primary localized alveolar rhabdomyosarcoma (RMA) are common. Post-relapse survival is poor. We evaluated prognostic factors including relapse treatment in patients with recurrent RMA. METHODS: Relapses occurred in 115/235 patients with nonmetastatic RMA treated in four consecutive CWS-trials after achievement of a complete remission. Sufficient information about post-relapse treatment and outcome could be obtained in 99 patients and was retrospectively analyzed. RESULTS: Nine of 99 patients received no salvage therapy and died after a median of 2 months. The remaining 90 patients received multimodal relapse treatment including mandatory chemotherapy. Recurrences were grossly resected in 39 patients; 57 patients received radiation. At a median follow-up from relapse of 8 years, 20 patients were alive and disease-free (5-year post-relapse survival [PROS] 21.3 ± 8). All surviving patients apart from a single individual had an isolated, circumscribed recurrence. Sixteen of 20 survivors were treated with adequate local relapse therapy (ALRT, i.e., either complete resection or gross resection + radiation). Survival in the subgroup of 27 individuals with circumscribed recurrences and ALRT was significantly better (PROS 53.7 ± 19) compared with disseminated recurrences and/or tumors treated without ALRT. Absence of primary lymph node involvement, circumscribed relapses, ALRT, and achievement of a second CR were identified as independent favorable risk factors. CONCLUSION: Post-relapse survival for primary localized RMA is generally poor. However, certain patient groups differed significantly in their likelihood of survival and 50% of patients with circumscribed relapses treated with ALRT survived. These findings may form the basis for an evidence-based risk-stratification for recurrent disease including relapse treatment.

Treatment efficiency, outcome and surgical treatment problems in patients suffering from localized embryonal bladder/prostate rhabdomyosarcoma: a report from the Cooperative Soft Tissue Sarcoma trial CWS-96.

BACKGROUND: To analyze the clinical course, treatment modalities, complications and outcome of patients suffering from localized embryonal bladder/prostate rhabdomyosarcoma (BPRMS) treated on the CWS-96 trial. PROCEDURE: There were 85 patients with BPRMS enrolled and 63 patients with embryonal non-metastatic BPRMS were analyzed. Fifty-six patients received neoadjuvant chemotherapy and response was assessed radiographically after 9 weeks. Local therapy with radiation and or surgery was performed based on age, tumor size, and response. Patients were treated with adjuvant chemotherapy following local control. RESULTS: Patient's age ranged from 0 to 16 years with a median follow up of 5.3 years. Eighty nine percent of the patients had IRS group III disease. The 5-year overall survival (OS) for the whole group was 76.3 ± 5.6% and the 5-year event-free survival (EFS) 69.8 ± 6.2%. Seventeen patients underwent preoperative radiochemotherapy followed by tumor resection (5-year-OS: 87.8 ± 8.1%). Eight patients were treated with solely radiochemotherapy (87.5 ± 11.7%). Twenty-five patients received chemotherapy and tumor resection (OS: 83.6 ± 7.5%). Thirteen patients underwent incomplete tumor resection and were treated with radiochemotherapy postoperatively (OS: 39.9 ± 14.8%, P < 0.05 vs. other groups). CONCLUSIONS: Local therapy is an important factor for prognosis of localized embryonal BPRMS. Inadequate primary or secondary surgery compromises the outcome and should be avoided. Radiotherapy alone, complete surgical tumor resection or combined preoperative radiotherapy with surgical resection lead to similar good local control rates and prognosis.

Cooperative trial CWS-91 for localized soft tissue sarcoma in children, adolescents, and young adults.

PURPOSE: To improve risk-adapted therapy for localized childhood soft tissue sarcoma within an international multicenter setting. PATIENTS AND METHODS: Four hundred forty-one patients younger than 21 years with localized rhabdomyosarcoma and rhabdomyosarcoma-like tumors (ie, extraosseous tumors of the Ewing family, synovial sarcoma, and undifferentiated sarcoma) were eligible. Therapy was stratified according to postsurgical stage, histology, and tumor site. In unresectable tumors, treatment was further adapted depending on response to induction chemotherapy, TN classification, tumor size and second-look surgery. A novel five-drug combination of etoposide, vincristine, dactinomycin, ifosfamide, and doxorubicin (EVAIA) was evaluated for high-risk patients, but cumulative chemotherapy dosage and treatment duration were reduced for the remaining individuals as compared with that of the previous trial CWS-86. Hyperfractionated accelerated radiotherapy (HART) was recommended at doses of either 32 or 48 Gy. RESULTS: At a median follow-up of 8 years, 5-year event-free survival (EFS) and overall (OS) survival for the entire cohort was 63% +/- 4% and 73% +/- 4%, respectively (all survival rates in this abstract are calculated and displayed with +/-95% CI). EFS/OS rates by histology were 60% +/- 5%/72% +/- 5% in rhabdomyosarcoma, 62% +/- 10%/69% +/- 10% for Ewing tumors of soft tissues, 84% +/- 12%/90% +/- 10% for synovial sarcoma, and 67% +/- 38%/83% +/- 30% for undifferentiated sarcoma, respectively. Response to one cycle of the five-drug combination EVAIA was similar to that of the four-drug combination VAIA used in CWS-86. Two hundred twelve patients with rhabdomyosarcoma underwent radiation (EFS, 66% +/- 6%); 53 of those patients had a favorable risk profile and received 32 Gy of HART (EFS, 73% +/- 12%). TN classification, tumor site, tumor size, histology, and age were prognostic in univariate analysis. CONCLUSION: Improved risk stratification enabled decreased therapy intensity for selected patients without compromising survival. Intensified chemotherapy with EVAIA did not improve outcome of localized high-risk rhabdomyosarcoma.

Mesenchymal chondrosarcoma of soft tissues and bone in children, adolescents, and young adults: experiences of the CWS and COSS study groups.

BACKGROUND: Mesenchymal chondrosarcoma (MCS) is a rare tumor with a strong tendency toward late recurrences leading to reported 10-year survival rates below 50%. The recommended treatment is resection with wide margins; the effectiveness of chemo- and radiotherapy remain poorly defined. As reports about MCS in young patients are scarce, treatment and outcomes of children/adolescents/young adults in the CWS and COSS studies were investigated. METHODS: Since 1977, 15 of >7000 CWS and COSS patients

Treatment of children with metastatic soft tissue sarcoma with oral maintenance compared to high dose chemotherapy: report of the HD CWS-96 trial.

PURPOSE: We prospectively studied the efficacy of high dose therapy (HDT) versus an oral maintenance treatment (OMT) in patients with stage IV soft tissue sarcoma (STS). PATIENTS AND METHODS: Both groups were pretreated with the CEVAIE combination consisting of carboplatin, etoposide, vincristine, actinomycin D, ifosfamide, and epirubicin. HDT consisted of a tandem cycle of thiotepa (600 mg/m(2)) plus cyclophosphamide (4,500 mg/m(2)) and melphalan (120 mg/m(2)) plus etoposide (1,800 mg/m(2)). This treatment was compared with OMT, consisting of four cycles trofosfamide (10 days 2 x 75 mg/m(2)/day) plus etoposide (10 days 2 x 25 mg/m(2)/day), and 4 cycles trofosfamide (10 days 2 x 75 mg/m(2)/day) plus idarubicin (10 days 4 x 5 mg/m(2)). Eligibility criteria were: diagnosis confirmed by reference pathology, primary stage IV, below 22 years of age, and having completed the study therapy. RESULTS: From 96 patients 45 were treated with HDT and 51 with OMT. The main risk parameters were equally distributed in both arms. After a median follow-up of 57.4 months, 11/45 (24.4%) patients in the HDT-arm and 26/51 (57.8%) patients in OMT-arm were alive. Kaplan-Meier analysis demonstrated an overall survival for the whole group of 0.27 (OMT group: 0.52, HDT group 0.27, log rank P = 0.03). The proportional hazard analysis for patients with rhabdomyosarcoma (RMS) or "RMS-like" tumors (77.1% of all patients) demonstrated an independent benefit of OMT on outcome. CONCLUSION: Oral maintenance therapy seems to be a promising option for patients with RMS-like stage IV tumors.

Initial patient characteristics can predict pattern and risk of relapse in localized rhabdomyosarcoma.

PURPOSE: Evaluation of primary tumor-, treatment-, and patient-related factors predicting relapse pattern, risk, and survival after relapse with the aim to design a risk-adapted, tumor-directed surveillance program for patients with localized rhabdomyosarcoma (RMS). PATIENTS AND METHODS: One thousand one hundred sixty-four patients with nonmetastatic RMS achieved complete remission at the end of multimodal therapy in the consecutive trials of the Cooperative Weichteilsarkom Studiengruppe (CWS)-81, CWS-86, CWS-91, and CWS-96 between 1980 and 2002 (median follow-up, 5 years). Three hundred thirty-seven of these individuals developed either locoregional, metastatic, or combined relapses. Predictive factors for relapse, its pattern, and postrelapse survival were analyzed. RESULTS: Age, histology, tumor size, tumor site, postsurgical stage, and omission of radiotherapy were identified as factors associated with an increased relapse risk in multivariate analyses. Relapse rates did not differ among the CWS trials. Median time to relapse was 1.43 years from first diagnosis (range, 0.13 to 13.5 years). There were 217 locoregional, 72 metastatic, and 48 combined recurrences. Only two patients developed metastases more than 4 years after diagnosis, and both had combined recurrences. Five-year postrelapse survival was 24%. Patient subsets with consistent relapse pattern, risk, and postrelapse survival rates were identified on the basis of histologic subtype and tumor size. CONCLUSION: Initial patient and tumor characteristics predict pattern and risk of relapse and also correlate with postrelapse survival probabilities. In localized RMS, tumor-directed follow-up should focus on the primary site. Screening for metastatic relapse may not be necessary more than 4 years after diagnosis. The identification of subgroups with distinctive pattern and risk of relapse may be used to develop risk-adapted, tumor-directed guidance for detection of recurrent disease in localized RMS.

Grossly-resected synovial sarcoma treated by the German and Italian Pediatric Soft Tissue Sarcoma Cooperative Groups: discussion on the role of adjuvant therapies.

Differently from adult oncologists that considered synovial sarcoma (SS) a tumor with uncertain chemosensitiveness, since two decades pediatric oncologists in Europe assumed that chemotherapy played an important role in SS treatment, so most pediatric patients were included in rhabdomyosarcoma protocols, receiving adjuvant chemotherapy regardless of risk factors. The German and Italian groups reviewed the data of grossly resected SS patients in order to define a risk-adapted treatment program for the next European protocol. A total of 150 patients < 21 years with localized SS who underwent initial gross resection between 1975 and 2002 were the object of this study. All but four cases received adjuvant chemotherapy. Post-operative radiotherapy was administered to 50% Group I and to 92% Group II patients. Five-year event-free survival (EFS) and overall survival (OS) were 77% and 89%, respectively. Survival rates were influenced by tumor size (EFS 92% and 56% for size < or = and > 5 cm, respectively) and local invasiveness, not by surgical margins. No metastatic relapses occurred in Group I < or = 5 cm patients, while the outcome was poor for T2B patients (EFS 41%) due to a high rate of metastatic relapse. Our study was unable to assess the role of adjuvant treatments in grossly-resected SS, but identified a subset of low-risk patients (IRS Group I, size < or = 5 cm), for which the omission of adjuvant chemotherapy could be suggested, and a subset of high-risk patients (T2B), who need treatment intensification.

The role of adjuvant chemotherapy in children and adolescents with surgically resected, high-risk adult-type soft tissue sarcomas.

PURPOSE: This analysis evaluates whether adjuvant chemotherapy can be recommended for high-risk, surgically-resected, adult-type non-rhabdomyosarcoma soft tissue sarcomas (NRSTS) within the new European Pediatric Soft Tissue Sarcoma Study Group (EpSSG) protocol. The Italian and German Cooperative Groups reviewed their data-bases, analyzing patients classified as group I-II, with high-grade tumor (G3) larger than 5 cm in size. METHODS: The analysis included 36 patients, and compared the clinical features and outcome of the group of 21 patients who received chemotherapy versus the group of 15 patients treated with local therapies only. RESULTS: For the series as a whole, 5-year event-free survival (EFS), metastasis-free survival (MFS), and overall survival (OS) were 26.2%, 34.0%, and 37.5%, respectively. In patients treated with chemotherapy, MFS and OS were 49.5% and 41.5% (median time to relapse: 13 months). In patients who did not receive chemotherapy, MFS and OS were 0% and 23.8% (median time to relapse: 3 months). CONCLUSION: The role of adjuvant chemotherapy in NRSTS is still uncertain, however, the current retrospective analysis showed that: (1) despite the globally good prognosis of grossly-resected cases, patients with G3 and large-size have a high-risk of metastatic spread, and (2) MFS appears to be better in patients who had chemotherapy. Based in part on these results, and in accordance with recent suggestions coming from the literature on adult sarcomas, the EpSSG NRSTS protocol will recommend adjuvant chemotherapy in high-risk surgically-resected patients.