Abstract Purpose One of the major hurdles in enabling personalized medicine is obtaining sufficient patient data to feed into predictive models. Combining data originating from multiple hospitals is ...difficult because of ethical, legal, political, and administrative barriers associated with data sharing. In order to avoid these issues, a distributed learning approach can be used. Distributed learning is defined as learning from data without the data leaving the hospital. Patients and methods Clinical data from 287 lung cancer patients, treated with curative intent with chemoradiation (CRT) or radiotherapy (RT) alone were collected from and stored in 5 different medical institutes (123 patients at MAASTRO (Netherlands, Dutch), 24 at Jessa (Belgium, Dutch), 34 at Liege (Belgium, Dutch and French), 48 at Aachen (Germany, German) and 58 at Eindhoven (Netherlands, Dutch)). A Bayesian network model is adapted for distributed learning (watch the animation: http://youtu.be/nQpqMIuHyOk ). The model predicts dyspnea, which is a common side effect after radiotherapy treatment of lung cancer. Results We show that it is possible to use the distributed learning approach to train a Bayesian network model on patient data originating from multiple hospitals without these data leaving the individual hospital. The AUC of the model is 0.61 (95%CI, 0.51–0.70) on a 5-fold cross-validation and ranges from 0.59 to 0.71 on external validation sets. Conclusion Distributed learning can allow the learning of predictive models on data originating from multiple hospitals while avoiding many of the data sharing barriers. Furthermore, the distributed learning approach can be used to extract and employ knowledge from routine patient data from multiple hospitals while being compliant to the various national and European privacy laws.
Display omitted
•Developed and implemented IT infrastructure in 5 radiation clinics across 3 countries.•Proof-of-principle for ‘big data’ infrastructure and distributed learning studies.•General ...framework to execute learning algorithms on distributed data.
Machine learning applications for personalized medicine are highly dependent on access to sufficient data. For personalized radiation oncology, datasets representing the variation in the entire cancer patient population need to be acquired and used to learn prediction models. Ethical and legal boundaries to ensure data privacy hamper collaboration between research institutes. We hypothesize that data sharing is possible without identifiable patient data leaving the radiation clinics and that building machine learning applications on distributed datasets is feasible.
We developed and implemented an IT infrastructure in five radiation clinics across three countries (Belgium, Germany, and The Netherlands). We present here a proof-of-principle for future ‘big data’ infrastructures and distributed learning studies. Lung cancer patient data was collected in all five locations and stored in local databases. Exemplary support vector machine (SVM) models were learned using the Alternating Direction Method of Multipliers (ADMM) from the distributed databases to predict post-radiotherapy dyspnea grade ⩾2. The discriminative performance was assessed by the area under the curve (AUC) in a five-fold cross-validation (learning on four sites and validating on the fifth). The performance of the distributed learning algorithm was compared to centralized learning where datasets of all institutes are jointly analyzed.
The euroCAT infrastructure has been successfully implemented in five radiation clinics across three countries. SVM models can be learned on data distributed over all five clinics. Furthermore, the infrastructure provides a general framework to execute learning algorithms on distributed data. The ongoing expansion of the euroCAT network will facilitate machine learning in radiation oncology. The resulting access to larger datasets with sufficient variation will pave the way for generalizable prediction models and personalized medicine.
Abstract Purpose Radiotherapy-induced moist desquamation (RIMD) is a complication that can affect patients’ quality of life and jeopardize radiotherapy outcomes. The curative use of a hydroactive ...colloid gel has previously been shown effective in the management of RIMD in breast cancer patients. This study aimed at investigating the efficacy of this same gel but in the prevention of RIMD. Methods A group of breast cancer patients who applied the hydroactive gel from start to end of post-lumpectomy radiotherapy (Preventive Hydrogel group) were compared with two groups of matched historical controls: a group applying a dexpanthenol cream throughout their therapy and a group applying first the dexpanthenol cream then, after 11–14 fractions of radiotherapy, the hydroactive gel (Curative Hydrogel group). All patients received identical fractionation regimen. The clinical outcomes were the incidence and time to onset of RIMD. Key results After 25 fractions of radiotherapy (50 Gy), patients in the Preventive Hydrogel group (N = 202) developed RIMD significantly less frequently and later than patients in the Dexpanthenol group (N = 131; incidence = 7% vs 35% respectively, odds ratios = 7.27; probability of RIMD-free survival after 50 Gy = 0.88 vs 0.62). There were no significant differences between the Preventive and the Curative Hydrogel group (N = 87). Conclusions These findings confirm our previous results: applying the hydroactive colloid gel, rather than dexpanthenol, delayed the onset and reduced the incidence of RIMD in breast cancer patients. However, applying the hydrogel preventively offered no statistically significant advantages over applying it curatively.
The Cancer Rehabilitation Evaluation System (CARES) is a quality of life (QOL) and needs assessment instrument of US origin that was developed in the 90's. Since November 2012 the copyright and user ...fee were abolished and the instrument became publicly available the present study aims to reinvestigate the psychometric properties of the CARES for the Flemish population in Belgium.
The CARES was translated into Flemish following a translation-back translation process. A sample of 192 cancer patients completed the CARES, concurrent measures, and questions on socio-demographic and medical data. Participants were asked to complete the CARES a second time 1 week later, followed by some questions on their experiences with the instrument. Internal consistency, test-retest reliability, content validity, construct validity, concurrent validity and feasibility of the CARES were subsequently assessed.
The Flemish CARES version demonstrated excellent reliability with high internal consistency (range .87-.96) and test-retest ratings (range .70-.91) for all summary scales. Factor analysis replicated the original factor solution of five higher order factors with factor loadings of .325-.851. Correlations with other instruments ranging from |.43|-|.75| confirmed concurrent validity. Feasibility was indicated by the low number of missing items (mean 2.3; SD 5.0) and positive feedback of participants on the instrument.
The Flemish CARES has strong psychometric properties and can as such be a valid tool to assess cancer patients' QOL and needs in research, for example in international comparisons. The positive feedback of participants on the CARES support the usefulness of this tool for systematic assessment of cancer patients' well-being and care needs in clinical practice.
ClinicalTrials.gov: NCT02282696 (July 16, 2014).
Celotno besedilo
Dostopno za:
DOBA, IZUM, KILJ, NUK, PILJ, PNG, SAZU, SIK, UILJ, UKNU, UL, UM, UPUK
The purpose of this paper is to analyse the utilization of formal and informal home care among older patients with cancer (OCP) and to compare this with middle-aged patients with cancer (MCP) and ...older patients without cancer (ONC). Additionally, we examined predictors of transitions towards formal care one year after a cancer diagnosis.
OCP and MCP had to be recruited within three months after a cancer diagnosis and have an estimated life expectancy over six months. ONC consisted of patients without known cancer, seen by the general practitioner. Formal and informal care were compared between the patient groups at baseline, i.e. shortly after a cancer diagnosis and changes in care were studied after one year.
A total of 844 patients were evaluable for formal care at baseline and 469 patients (56%) at follow-up. At baseline, about half of older adults and 18% of MCP used formal care, while about 85% of cancer patients and 57% ONC used informal care. Formal care increased for all groups after one year though not significantly in OCP. The amount of informal care only changed in MCP which decreased after one year. Cancer-related factors and changes in need factors predict a transition towards formal care after a cancer diagnosis.
A cancer diagnosis has a different impact on the use of formal and informal care than ageing as such. The first year after a cancer diagnosis is an important time to follow-up on the patients' needs for home care.
Celotno besedilo
Dostopno za:
CEKLJ, DOBA, IZUM, KILJ, NUK, PILJ, PNG, SAZU, SIK, UILJ, UKNU, UL, UM, UPUK
Rationale, aims, and objectives
The systematic assessment of cancer patients well‐being and care needs is internationally recommended to optimize comprehensive cancer care. The Cancer Rehabilitation ...Evaluation System (CARES) is a psychometrically robust quality of life and needs assessment tool of US origin, developed in the early 1990s. This article describes Belgian patients' view on the content validity and feasibility of the CARES for use in current cancer care.
Methods
Participants were cancer patients recruited through media. Data were gathered in 4 focus groups (n = 26). The focus group discussions were facilitated with key questions. A moderator and an observer conducted and followed up the discussion. The audio file was transcribed verbatim and afterwards analyzed thematically.
Results
Participants experience concerns and needs in a wide range of life domains such as physical, emotional, cognitive, social, relational, sexual, financial, and work‐related and in the interaction with care professionals. According to participants, the items of the CARES are all relevant to capture the possible life disruption that cancer patients and survivors experience. One important theme is missing in the CARES, namely, the well‐being of loved ones. The completion time of the CARES was judged to be feasible, and according to participants, only a few items need a reformulation.
Conclusions
In general, the results of this study support the content validity and feasibility of the CARES. However, little adjustments in formulation and a few extra items are needed. The instrument can be used to obtain a comprehensive assessment of a cancer patients' overall well‐being and care needs to take dedicated action in care.
The aim of this Letter to the Editor was to report some methodological shortcomings in the recently published article "Application of red light phototherapy in the treatment of radioactive dermatitis ...in patients with head and neck cancer" by Zhang et al. There are some issues regarding the incomplete photobiomodulation (PBM) parameters, the chosen outcome measures, and some missing reference articles. In conclusion, the results of this study should be interpreted with caution and further research is necessary.
Breast cancer is the most common cancer in women worldwide, with an incidence of 1.7 million in 2012. Breast cancer and its treatments can bring along serious side effects such as fatigue, skin ...toxicity, lymphedema, pain, nausea, etc. These can substantially affect the patients’ quality of life. Therefore, supportive care for breast cancer patients is an essential mainstay in the treatment. Low-level light therapy (LLLT) also named photobiomodulation therapy (PBMT) has proven its efficiency in general medicine for already more than 40 years. It is a noninvasive treatment option used to stimulate wound healing and reduce inflammation, edema, and pain. LLLT is used in different medical settings ranging from dermatology, physiotherapy, and neurology to dentistry. Since the last twenty years, LLLT is becoming a new treatment modality in supportive care for breast cancer. For this review, all existing literature concerning the use of LLLT for breast cancer was used to provide evidence in the following domains: oral mucositis (OM), radiodermatitis (RD), lymphedema, chemotherapy-induced peripheral neuropathy (CIPN), and osteonecrosis of the jaw (ONJ). The findings of this review suggest that LLLT is a promising option for the management of breast cancer treatment-related side effects. However, it still remains important to define appropriate treatment and irradiation parameters for each condition in order to ensure the effectiveness of LLLT.
Purpose
The aim of this study was to assess the effectiveness and acceptability of photobiomodulation using MLS
®
laser therapy (LT) in the management of acute radiation dermatitis (RD).
Methods
We ...compared two successive groups of breast cancer patients undergoing identical radiotherapy regimens post-lumpectomy. Both groups received our standard skin care but the second group received six additional LT sessions (beam area 19.635 cm
2
, 0.168 W/cm
2
, 4 J/cm
2
), starting at fraction 20 of radiotherapy (control and LT group,
N
= 41 and 38, respectively). The clinical outcomes were the severity of RD (using the Radiation Therapy Oncology Group RTOG criteria and the Radiotherapy-Induced Skin Reaction Assessment Scale RISRAS) and dermatology-specific quality of life (Skindex-16) before the start of LT and at the end of radiotherapy. Secondary outcomes were patients’ ratings of skin care or LT (pleasantness, soothing effect, and global satisfaction).
Results
Skin toxicity was equivalent between the groups before the start of LT but significantly differed at the end of radiotherapy, with an aggravation in the control but not in the LT group (e.g., 29 versus 3 % of RTOG grade 2 RD, respectively,
P
< 0.005). We found no significant group differences with respect to quality of life. However, the RISRAS subjective score decreased in the LT group only, implying a decreased impact of RD on patients’ quality of life. Finally, patients’ ratings were significantly higher for LT than for standard care.
Conclusions
These findings suggest that LT might be effective to manage acute RD and warrant further research.
Trial registration
Clinical trial number: NCT01932073.
https://clinicaltrials.gov/ct2/show/NCT01932073
.
Cancer is mainly a disease of older patients. In older cancer patients, additional endpoints such as quality of survival and daily functioning might be considered equally relevant as overall or ...disease free survival. However, these factors have been understudied using prospective designs focussing on older cancer patients. Therefore, this study will focus on the impact of cancer, ageing, and their interaction on the long-term wellbeing of older cancer patients.
This study is an observational cohort study. We aim to recruit 720 cancer patients above 70 years with a new diagnosis of breast, prostate, lung or gastrointestinal cancer and two control groups: one control group of 720 patients above 70 years without a previous diagnosis of cancer and one control group of 720 cancer patients between 50 - 69 years newly diagnosed with breast, prostate, lung or gastrointestinal cancer. Data collection will take place at inclusion, after six months, after one year and every subsequent year until death or end of the study. Data will be collected through personal interviews (consisting of socio-demographic information, general health information, a comprehensive geriatric assessment, quality of life, health locus of control and a loneliness scale), a handgrip test, assessment of medical records, two buccal swabs and a blood sample from cancer patients (at baseline). As an annex study, caregivers of the participants will be recruited as well. Data collection for caregivers will consist of a self-administered questionnaire examining depression, coping, and burden.
This extensive data collection will increase insight on how wellbeing of older cancer patients is affected by cancer (diagnosis and treatment), ageing, and their interaction. Results may provide new insights, which might contribute to the improvement of care for older cancer patients.
Celotno besedilo
Dostopno za:
DOBA, IZUM, KILJ, NUK, PILJ, PNG, SAZU, SIK, UILJ, UKNU, UL, UM, UPUK