Machine learning-based biomarker profile derived from 4210 serially measured proteins predicts clinical outcome of patients with heart failure

Aims

Risk assessment tools are needed for timely identification of patients with heart failure (HF) with reduced ejection fraction (HFrEF) who are at high risk of adverse events. In this study, we aim to derive a small set out of 4210 repeatedly measured proteins, which, along with clinical characteristics and established biomarkers, carry optimal prognostic capacity for adverse events, in patients with HFrEF.

Methods and results

In 382 patients, we performed repeated blood sampling (median follow-up: 2.1 years) and applied an aptamer-based multiplex proteomic approach. We used machine learning to select the optimal set of predictors for the primary endpoint (PEP: composite of cardiovascular death, heart transplantation, left ventricular assist device implantation, and HF hospitalization). The association between repeated measures of selected proteins and PEP was investigated by multivariable joint models. Internal validation (cross-validated c-index) and external validation (Henry Ford HF PharmacoGenomic Registry cohort) were performed. Nine proteins were selected in addition to the MAGGIC risk score, N-terminal pro-hormone B-type natriuretic peptide, and troponin T: suppression of tumourigenicity 2, tryptophanyl-tRNA synthetase cytoplasmic, histone H2A Type 3, angiotensinogen, deltex-1, thrombospondin-4, ADAMTS-like protein 2, anthrax toxin receptor 1, and cathepsin D. N-terminal pro-hormone B-type natriuretic peptide and angiotensinogen showed the strongest associations [hazard ratio (95% confidence interval): 1.96 (1.17-3.40) and 0.66 (0.49-0.88), respectively]. The multivariable model yielded a c-index of 0.85 upon internal validation and c-indices up to 0.80 upon external validation. The c-index was higher than that of a model containing established risk factors (P = 0.021).

Conclusion

Nine serially measured proteins captured the most essential prognostic information for the occurrence of adverse events in patients with HFrEF, and provided incremental value for HF prognostication beyond established risk factors. These proteins could be used for dynamic, individual risk assessment in a prospective setting. These findings also illustrate the potential value of relatively 'novel' biomarkers for prognostication.

Clinical Trial Registration

https://clinicaltrials.gov/ct2/show/NCT01851538?term=nCT01851538&draw=2&rank=1 24.

Dynamic Digital Twin: Diagnosis, Treatment, Prediction, and Prevention of Disease During the Life Course

A digital twin (DT), originally defined as a virtual representation of a physical asset, system, or process, is a new concept in health care. A DT in health care is not a single technology but a domain-adapted multimodal modeling approach incorporating the acquisition, management, analysis, prediction, and interpretation of data, aiming to improve medical decision-making. However, there are many challenges and barriers that must be overcome before a DT can be used in health care. In this viewpoint paper, we build on the current literature, address these challenges, and describe a dynamic DT in health care for optimizing individual patient health care journeys, specifically for women at risk for cardiovascular complications in the preconception and pregnancy periods and across the life course. We describe how we can commit multiple domains to developing this DT. With our cross-domain definition of the DT, we aim to define future goals, trade-offs, and methods that will guide the development of the dynamic DT and implementation strategies in health care.

The role of perioperative chemotherapy in primary high-grade extremity soft tissue sarcoma: a risk-stratified analysis using PERSARC

OBJECTIVE

The aim of the study is to assess the effect of perioperative chemotherapy (CTX) in patients with grade II-III extremity soft tissue sarcoma (eSTS) on overall survival (OS) and evaluate whether the PERSARC prediction tool could identify patients with eSTS more likely to benefit from CTX.

METHODS

Patients (18-70 years) with primary high-grade eSTS surgically treated with curative intent were included in the retrospective cohort study. The effect of any perioperative CTX and anthracycline + ifosfamide (AI)-based CTX on OS was investigated in three PERSARC-risk groups (high/intermediate/low). The PERSARC-risk groups were defined by the 33% and 66% quantile of the predicted 5-year OS of the study population equal to a 5-year OS of 65.8% and 79.8%, respectively. The effect of CTX on OS was investigated with weighted Kaplan-Meier curves and multivariable Cox models with an interaction between risk group and CTX.

RESULTS

This study included 5683 patients. The weighted Kaplan-Meier curves did not demonstrate a beneficial effect of any CTX and AI-based CTX on OS in the overall population. However, in the high PERSARC-risk group the 5-year OS of AI-based CTX was significantly better than no CTX (69.8% vs 59.0%, respectively, p = 0.004) (HR 0.66, 95%CI 0.53-0.83).

CONCLUSIONS

This study demonstrated a beneficial effect of AI-based CTX on OS in a selected group of high-risk patients with an absolute survival benefit of 11% as stratified by the PERSARC prediction tool. However, no beneficial effect of CTX on OS was found in the overall population of patients with primary high-grade eSTS younger than 70 years.

Survival Analysis of 3 Different Age Groups and Prognostic Factors among 402 Patients with Skeletal High-Grade Osteosarcoma. Real World Data from a Single Tertiary Sarcoma Center

Age is a known prognostic factor for many sarcoma subtypes, however in the literature there are limited data on the different risk profiles of different age groups for osteosarcoma survival. This study aims to provide an overview of survival in patients with high-grade osteosarcoma in different age groups and prognostic variables for survival and local control among the entire cohort. In this single center retrospective cohort study, 402 patients with skeletal high-grade osteosarcoma were diagnosed and treated with curative intent between 1978 and 2017 at the Leiden University Medical Center (LUMC). Prognostic factors for survival were analyzed using a Cox proportional hazard model. In this study poor overall survival (OS) and event-free survival (EFS) were associated with increasing age. Age groups, tumor size, poor histopathological response, distant metastasis (DM) at presentation and local recurrence (LR) were important independent prognostic factors influencing OS and EFS. Differences in outcome among different age groups can be partially explained by patient and treatment characteristics.

External validation and adaptation of a dynamic prediction model for patients with high-grade extremity soft tissue sarcoma

BACKGROUND AND OBJECTIVES

A dynamic prediction model for patients with soft tissue sarcoma of the extremities was previously developed to predict updated overall survival probabilities throughout patient follow-up. This study updates and externally validates the dynamic model.

METHODS

Data from 3826 patients with high-grade extremity soft tissue sarcoma, treated surgically with curative intent were used to update the dynamic PERsonalised SARcoma Care (PERSARC) model. Patients were added to the model development cohort and grade was included in the model. External validation was performed with data from 1111 patients treated at a single tertiary center.

RESULTS

Calibration plots show good model calibration. Dynamic C-indices suggest that the model can discriminate between high- and low-risk patients. The dynamic C-indices at 0, 1, 2, 3, 4, and 5 years after surgery were equal to 0.697, 0.790, 0.822, 0.818, 0.812, and 0.827, respectively.

CONCLUSION

Results from the external validation show that the dynamic PERSARC model is reliable in predicting the probability of surviving an additional 5 years from a specific prediction time point during follow-up. The model combines patient-, treatment-specific and time-dependent variables such as local recurrence and distant metastasis to provide accurate survival predictions throughout follow-up and is available through the PERSARC app.

Dynamic prediction of overall survival: a retrospective analysis on 979 patients with Ewing sarcoma from the German registry

OBJECTIVES

This study aimed at developing a dynamic prediction model for patients with Ewing sarcoma (ES) to provide predictions at different follow-up times. During follow-up, disease-related information becomes available, which has an impact on a patient's prognosis. Many prediction models include predictors available at baseline and do not consider the evolution of disease over time.

SETTING

In the analysis, 979 patients with ES from the Gesellschaft für Pädiatrische Onkologie und Hämatologie registry, who underwent surgery and treatment between 1999 and 2009, were included.

DESIGN

A dynamic prediction model was developed to predict updated 5-year survival probabilities from different prediction time points during follow-up. Time-dependent variables, such as local recurrence (LR) and distant metastasis (DM), as well as covariates measured at baseline, were included in the model. The time effects of covariates were investigated by using interaction terms between each variable and time.

RESULTS

Developing LR, DM in the lungs (DMp) or extrapulmonary DM (DMo) has a strong effect on the probability of surviving an additional 5 years with HRs and 95% CIs equal to 20.881 (14.365 to 30.353), 6.759 (4.465 to 10.230) and 17.532 (13.210 to 23.268), respectively. The effects of primary tumour location, postoperative radiotherapy (PORT), histological response and disease extent at diagnosis on survival were found to change over time. The HR of PORT versus no PORT at the time of surgery is equal to 0.774 (0.594 to 1.008). One year after surgery, the HR is equal to 1.091 (0.851 to 1.397).

CONCLUSIONS

The time-varying effects of several baseline variables, as well as the strong impact of time-dependent variables, show the importance of including updated information collected during follow-up in the prediction model to provide accurate predictions of survival.

Surgical Outcome and Oncological Survival of Osteofibrous Dysplasia-Like and Classic Adamantinomas: An International Multicenter Study of 318 Cases

BACKGROUND

Osteofibrous dysplasia-like adamantinoma (OFD-AD) and classic adamantinoma (AD) are rare, neoplastic diseases with only limited data supporting current treatment protocols. We believe that our retrospective multicenter cohort study is the largest analysis of patients with adamantinoma to date. The primary purpose of this study was to describe the disease characteristics and evaluate the oncological outcomes. The secondary purpose was to identify risk factors for local recurrence after surgical treatment and propose treatment guidelines.

METHODS

Three hundred and eighteen confirmed cases of OFD-AD and AD for which primary treatment was carried out between 1985 and 2015 were submitted by 22 tertiary bone tumor centers. Proposed clinical risk factors for local recurrence such as size, type, and margins were analyzed using univariable and multivariate Cox regression analysis.

RESULTS

Of the 318 cases, 128 were OFD-AD and 190 were AD. The mean age at diagnosis was 17 years (median, 14.5 years) for OFD-AD and 32 years (median, 28 years) for AD; 53% of the patients were female. The mean tumor size in the OFD-AD and AD groups combined was 7.8 cm, measured histologically. Sixteen percent of the patients sustained a pathological fracture prior to treatment. Local recurrence was recorded in 22% of the OFD-AD cases and 24% of the AD cases. None of the recurrences in the OFD-AD group progressed to AD. Metastatic disease was found in 18% of the AD cases and fatal disease, in 11% of the AD cases. No metastatic or fatal disease was reported in the OFD-AD group. Multivariate Cox regression analysis demonstrated that uncontaminated resection margins (hazard ratio [HR] = 0.164, 95% confidence interval [CI] = 0.092 to 0.290, p < 0.001), pathological fracture (HR = 1.968, 95% CI = 1.076 to 3.600, p = 0.028), and sex (female versus male: HR = 0.535, 95% CI = 0.300 to 0.952, p = 0.033) impacted the risk of local recurrence.

CONCLUSIONS

OFD-AD and AD are parts of a disease spectrum but should be regarded as different entities. Our results support reclassification of OFD-AD into the intermediate locally aggressive category, based on the local recurrence rate of 22% and absence of metastases. In our study, metastatic disease was restricted to the AD group (an 18% rate). We advocate wide resection with uncontaminated margins including bone and involved periosteum for both OFD-AD and AD.

LEVEL OF EVIDENCE

Prognostic Level IV. See Instructions for Authors for a complete description of levels of evidence.

Individual risk evaluation for local recurrence and distant metastasis in Ewing sarcoma: A multistate model: A multistate model for Ewing sarcoma

BACKGROUND

We investigated the effects of surgical margins, histological response, and radiotherapy on local recurrence (LR), distant metastasis (DM), and survival in Ewing sarcoma.

PROCEDURE

Disease evolution was retrospectively studied in 982 patients with Ewing sarcoma undergoing surgery after chemotherapy using a multistate model with initial state surgery, intermediate states LR, pulmonary metastasis (DMpulm), other DM ± LR (DMother), and final state death. Effect of risk factors was estimated using Cox proportional hazard models.

RESULTS

The median follow-up was 7.6 years (95% CI, 7.2-8.0). Risk factors for LR are pelvic location, HR 2.04 (1.10-3.80), marginal/intralesional resection, HR 2.28 (1.25-4.16), and radiotherapy, HR 0.52 (0.28-0.95); for DMpulm the risk factors are <90% necrosis, HR 2.13 (1.13-4.00), and previous pulmonary metastasis, HR 4.90 (2.28-8.52); for DMother are 90% to 99% necrosis, HR 1.56 (1.09-2.23), <90% necrosis, HR 2.66 (1.87-3.79), previous bone/other metastasis, HR 3.08 (2.03-4.70); and risk factors for death without LR/DM are pulmonary metastasis, HR 8.08 (4.01-16.29), bone/other metastasis, HR 10.23 (4.90-21.36), and <90% necrosis, HR 6.35 (3.18-12.69). Early LR (0-24 months) negatively influences survival, HR 3.79 (1.34-10.76). Once DMpulm/DMother arise only previous bone/other metastasis remain prognostic for death, HR 1.74 (1.10-2.75).

CONCLUSION

Disease extent and histological response are risk factors for progression to DM or death. Tumor site and surgical margins are risk factors for LR. If disease progression occurs, previous risk factors lose their relevance. In case of isolated LR, time to recurrence is important for decision-making. Radiotherapy seems protective for LR especially in pelvic/axial. Low percentages of LR in extremity tumors and associated toxicity question the need for radiotherapy in extremity Ewing sarcoma.

Surgical Treatment of Localized-Type Tenosynovial Giant Cell Tumors of Large Joints: A Study Based on a Multicenter-Pooled Database of 31 International Sarcoma Centers

BACKGROUND

Localized-type tenosynovial giant cell tumor (TGCT) is a rare, neoplastic disease with only limited data supporting treatment protocols. We describe treatment protocols and evaluate their oncological outcome, complications, and functional results in a large multicenter cohort of patients. A secondary study aim was to identify factors associated with local recurrence after surgical treatment.

METHODS

Patients with histologically proven localized TGCT of a large joint were included if they had been treated between 1990 and 2017 in 1 of 31 tertiary sarcoma centers. Of 941 patients with localized TGCT, 62% were female. The median age at initial treatment was 39 years, and the median duration of follow-up was 34 months. Sixty-seven percent of the tumors affected the knee, and the primary treatment at the tertiary center was 1-stage open resection in 73% of the patients. Proposed factors for predicting a first local recurrence after treatment in the tertiary center were tested in a univariate analysis, and those that demonstrated significance were subsequently included in a multivariate analysis.

RESULTS

The localized TGCT recurred in 12% of all cases, with local-recurrence-free rates at 3, 5, and 10 years of 88%, 83%, and 79%, respectively. The strongest factor for predicting recurrent disease was a prior recurrence (p < 0.001). Surgical treatment decreased pain and swelling in 71% and 85% of the patients, respectively, and such treatment was associated with complications in 4% of the patients. Univariate and multivariate analyses of the patients who had not undergone therapy previously yielded positive associations between local recurrence and a tumor size of ≥5 cm versus <5 cm (hazard ratio [HR] = 2.50; 95% confidence interval [CI] = 1.32 to 4.74; p = 0.005). Arthroscopy (versus open surgery) was significantly associated with tumor recurrence in the univariate analysis (p = 0.04) but not in the multivariate analysis (p = 0.056).

CONCLUSIONS

Factors associated with recurrence after resection of localized-type TGCT were larger tumor size and initial treatment with arthroscopy. Relatively low complication rates and good functional outcomes warrant an open approach with complete resection when possible to reduce recurrence rates in high-risk patients.

LEVEL OF EVIDENCE

Therapeutic Level III. See Instructions for Authors for a complete description of levels of evidence.

Surgical outcomes of patients with diffuse-type tenosynovial giant-cell tumours: an international, retrospective, cohort study

BACKGROUND

Diffuse-type tenosynovial giant-cell tumour is a rare, locally aggressive, and difficult-to-treat soft tissue tumour. Clinical and surgical outcomes depend on multiple factors, including preoperative diagnostic assessment, the localisation and extent of disease, and possibly the choice of treatment modalities by orthopaedic surgeons. We did a retrospective cohort study to characterise global surgical treatment protocols, and assess surgical outcomes, complications, and functional results in patients with diffuse-type tenosynovial giant-cell tumours.

METHODS

In this international, multicentre, retrospective cohort study, we included consecutive patients treated in 31 sarcoma reference centres between Jan 1, 1990, and Dec 31, 2017. Eligible patients were of any age and had histologically proven diffuse-type tenosynovial giant-cell tumour of large joints. Patient data were retrieved from the local databases of participating centres. Patients with localised-type tenosynovial giant-cell tumour were excluded. In the analysis, we only included patients with complete core criteria data regarding admission status, date of treatment, type of treatment at participating centre, and first local recurrence after treatment. We used a non-parametric method to estimate recurrence-free survival at 3, 5, and 10 years after initial surgical resection in a tertiary centre. We used a multivariate Cox regression model to estimate the effect of risk factors. We also present subgroup analyses of disease status at presentation (primary vs recurrent disease) and recurrence-free survival by surgery type (open surgery vs arthroscopic synovectomy), and prespecified risk factors were tested in a univariate and multivariable analyses, with an endpoint of first local recurrence after treatment in a tertiary centre.

FINDINGS

Data collection for these analyses occurred between January, 2016, and May, 2018. We received the records of 1192 patients, of which 966 (81%) were surgically treated and had complete information on core criteria. 445 patients were admitted with therapy-naive disease of the knee and were primarily treated in a tertiary centre. Since patients with wait and see treatment do not have a starting date of treatment, these patients were excluded in the calculation of median follow-up time for all patients. For this calculation we used time of surgery as a starting date. 758 (64%) of 1192 patients had knee involvement and 628 (54%) of 1163 patients with complete data on type of surgery had one-staged open synovectomy. At a median follow-up of 54 months (IQR 27-97), recurrent disease developed in 425 (44%) of all 966 surgically treated cases, and recurrence-free survival was 62% (95% CI 59-65) at 3 years, 55% (51-58) at 5 years, and 40% (35-45) at 10 years. Surgical complications were reported in 105 (12%) of 906 patients who had complete data on surgical complications. Pain improved after surgical treatment in 255 (59%) of 434 patients and swelling improved in 328 (72%) of 453 patients who had complete data.

INTERPRETATION

This study of patients with diffuse-type tenosynovial giant-cell tumour provides a comprehensive and up-to-date disease overview, assessing the clinical profile and management of the disease in multiple specialised referral centres. Surgical treatment of diffuse-type tenosynovial giant cell tumours is not a definitive treatment for every patient because it involves a high risk for local recurrent disease and a relatively high risk for postoperative complications. After surgical treatment in treatment-naive patients, risk factors for recurrent disease in individual patients were not identified in what we believe is the largest cohort to date.

FUNDING

Daiichi Sankyo.

Investigating hospital heterogeneity with a competing risks frailty model

Survival analysis is used in the medical field to identify the effect of predictive variables on time to a specific event. Generally, not all variation of survival time can be explained by observed covariates. The effect of unobserved variables on the risk of a patient is called frailty. In multicenter studies, the unobserved center effect can induce frailty on its patients, which can lead to selection bias over time when ignored. For this reason, it is common practice in multicenter studies to include a random frailty term modeling center effect. In a more complex event structure, more than one type of event is possible. Independent frailty variables representing center effect can be incorporated in the model for each competing event. However, in the medical context, events representing disease progression are likely related and correlation is missed when assuming frailties to be independent. In this work, an additive gamma frailty model to account for correlation between frailties in a competing risks model is proposed, to model frailties at center level. Correlation indicates a common center effect on both events and measures how closely the risks are related. Estimation of the model using the expectation‐maximization algorithm is illustrated. The model is applied to a data set from a multicenter clinical trial on breast cancer from the European Organisation for Research and Treatment of Cancer (EORTC trial 10854). Hospitals are compared by employing empirical Bayes estimates methodology together with corresponding confidence intervals.

Dynamic prediction of overall survival for patients with high-grade extremity soft tissue sarcoma

PURPOSE

There is increasing interest in personalized prediction of disease progression for soft tissue sarcoma patients. Currently, available prediction models are limited to predictions from time of surgery or diagnosis. This study updates predictions of overall survival at different times during follow-up by using the concept of dynamic prediction.

PATIENTS AND METHODS

Information from 2232 patients with high-grade extremity soft tissue sarcoma, who underwent surgery at 14 specialized sarcoma centers, was used to develop a dynamic prediction model. The model provides updated 5-year survival probabilities from different prediction time points during follow-up. Baseline covariates as well as time-dependent covariates, such as status of local recurrence and distant metastases, were included in the model. In addition, the effect of covariates over time was investigated and modelled accordingly in the prediction model.

RESULTS

Surgical margin and tumor histology show a significant time-varying effect on overall survival. The effect of margin is strongest shortly after surgery and diminishes slightly over time. Development of local recurrence and distant metastases during follow-up have a strong effect on overall survival and updated predictions must account for their occurrence.

CONCLUSION

The presence of time-varying effects, as well as the effect of local recurrence and distant metastases on survival, suggest the importance of updating predictions during follow-up. This newly developed dynamic prediction model which updates survival probabilities over time can be used to make better individualized treatment decisions based on a dynamic assessment of a patient's prognosis.

Individualised risk assessment for local recurrence and distant metastases in a retrospective transatlantic cohort of 687 patients with high-grade soft tissue sarcomas of the extremities: a multistate model

OBJECTIVES

This study investigates the effect of surgical margins and radiotherapy, in the presence of individual baseline characteristics, on survival in a large population of high-grade soft tissue sarcoma of the extremities using a multistate model.

DESIGN

A retrospective multicentre cohort study.

SETTING

4 tertiary referral centres for orthopaedic oncology.

PARTICIPANTS

687 patients with primary, non-disseminated, high-grade sarcoma only, receiving surgical treatment with curative intent between 2000 and 2010 were included.

MAIN OUTCOME MEASURES

The risk to progress from 'alive without disease' (ANED) after surgery to 'local recurrence' (LR) or 'distant metastasis (DM)/death'. The effect of surgical margins and (neo)adjuvant radiotherapy on LR and overall survival was evaluated taking patients' and tumour characteristics into account.

RESULTS

The multistate model underlined that wide surgical margins and the use of neoadjuvant radiotherapy decreased the risk of LR but have little effect on survival. The main prognostic risk factors for transition ANED to LR are tumour size (HR 1.06; 95% CI 1.01 to 1.11 (size in cm)) and (neo)adjuvant radiotherapy. The HRs for patients treated with adjuvant or no radiotherapy compared with neoadjuvant radiotherapy are equal to 4.36 (95% CI 1.34 to 14.24) and 14.20 (95% CI 4.14 to 48.75), respectively. Surgical resection margins had a protective effect for the occurrence of LR with HRs equal to 0.61 (95% CI 0.33 to 1.12), and 0.16 (95% CI 0.07 to 0.41) for margins between 0 and 2 mm and wider than 2 mm, respectively. For transition ANED to distant metastases/Death, age (HR 1.64 (95% CI 0.95 to 2.85) and 1.90 (95% CI 1.09 to 3.29) for 25-50 years and >50 years, respectively) and tumour size (1.06 (95% CI 1.04 to 1.08)) were prognostic factors.

CONCLUSIONS

This paper underlined the alternating effect of surgical margins and the use of neoadjuvant radiotherapy on oncological outcomes between patients with different baseline characteristics. The multistate model incorporates this essential information of a specific patient's history, tumour characteristics and adjuvant treatment modalities and allows a more comprehensive prediction of future events.