Fixing the Leaky Pipe: How to Improve the Uptake of Patient-Reported Outcomes-Based Prognostic and Predictive Models in Cancer Clinical Practice.

PURPOSE: This discussion paper outlines challenges and proposes solutions for successfully implementing prediction models that incorporate patient-reported outcomes (PROs) in cancer practice. METHODS: We organized a full-day multidisciplinary meeting of people with expertise in cancer care delivery, PRO collection, PRO use in prediction modeling, computing, implementation, and decision science. The discussions presented here focused on identifying challenges to the development, implementation and use of prediction models incorporating PROs, and suggesting possible solutions. RESULTS: Specific challenges and solutions were identified across three broad areas. (1) Understanding decision making and implementation: necessitating multidisciplinary collaboration in the early stages and throughout; early stakeholder engagement to define the decision problem and ensure acceptability of PROs in prediction; understanding patient/clinician interpretation of PRO predictions and uncertainty to optimize prediction impact; striving for model integration into existing electronic health records; and early regulatory alignment. (2) Recognizing the limitations to PRO collection and their impact on prediction: incorporating validated, clinically important PROs to maximize model generalizability and clinical engagement; and minimizing missing PRO data (resulting from both structural digital exclusion and time-varying factors) to avoid exacerbating existing inequalities. (3) Statistical and modeling challenges: incorporating statistical methods to address missing data; ensuring predictive modeling recognizes complex causal relationships; and considering temporal and geographic recalibration so that model predictions reflect the relevant population. CONCLUSION: Developing and implementing PRO-based prediction models in cancer care requires extensive multidisciplinary working from the earliest stages, recognition of implementation challenges because of PRO collection and model presentation, and robust statistical methods to manage missing data, causality, and calibration. Prediction models incorporating PROs should be viewed as complex interventions, with their development and impact assessment carried out to reflect this.

Quality of life support in advanced cancer-web and technological interventions: systematic review and narrative synthesis.

BACKGROUND: As treatments continue to progress, patients with advanced cancer are living longer. However, ongoing physical side-effects and psychosocial concerns can compromise quality of life (QoL). Patients and physicians increasingly look to the internet and other technologies to address diverse supportive needs encountered across this evolving cancer trajectory. OBJECTIVES: 1. To examine the features and delivery of web and technological interventions supporting patients with advanced cancer. 2. To explore their efficacy relating to QoL and psychosocial well-being. METHODS: Relevant studies were identified through electronic database searches (MEDLINE, PsychINFO, Embase, CINAHL, CENTRAL, Web of Science and ProQuest) and handsearching. Findings were collated and explored through narrative synthesis. RESULTS: Of 5274 identified records, 37 articles were included. Interventions were evaluated within studies targeting advanced cancer (13) or encompassing all stages (24). Five subtypes emerged: Interactive Health Communication Applications (n=12), virtual programmes of support (n=11), symptom monitoring tools (n=8), communication conduits (n=3) and information websites (n=3). Modes of delivery ranged from self-management to clinically integrated. Support largely targeted psychosocial well-being, alongside symptom management and healthy living. Most studies (78%) evidenced varying degrees of efficacy through QoL and psychosocial measures. Intervention complexity made it challenging to distinguish the most effective components. Incomplete reporting limited risk of bias assessment. CONCLUSION: While complex and varied in their content, features and delivery, most interventions led to improvements in QoL or psychosocial well-being across the cancer trajectory. Ongoing development and evaluation of such innovations should specifically target patients requiring longer-term support for later-stage cancer. PROSPERO REGISTRATION NUMBER: CRD42018089153.

Online Symptom Monitoring During Pelvic Radiation Therapy: Randomized Pilot Trial of the eRAPID Intervention.

PURPOSE: Radiation therapy (RT) and chemoRT for pelvic cancers increase survival but are associated with serious treatment-related symptoms. Electronic-patient self-Reporting of Adverse-events: Patient Information and aDvice (eRAPID) is a secure online system for patients to self-report symptoms, generating immediate advice for hospital contact or self-management. This pilot study aimed to establish feasibility and acceptability of the system. METHODS AND MATERIALS: In a prospective 2-center randomized parallel-group pilot study, patients undergoing radical pelvic RT for prostate cancer (prostateRT) or chemoRT for lower gastrointestinal and gynecological cancers were randomized to usual care (UC) or eRAPID (weekly online symptom reporting for 12, 18, and 24 weeks). Primary outcomes were recruitment/attrition, study completion, and patient adherence. Secondary outcomes were effect on hospital services and performance of patient outcome measures. Missing data, floor/ceiling effects, and mean change scores were examined for Functional Assessment of Cancer Therapy (FACT-G), European Organisation for Research and Treatment of Cancer, Quality of Life (EORTC QLQ C-30), self-efficacy, and EuroQol (EQ5D). RESULTS: From 228 patients approached, 167 (73.2%) were consented and randomized (83, eRAPID; 84, UC; 87, prostateRT; 80, chemoRT); 150 of 167 completed 24 study weeks. Only 16 patients (9.6%) withdrew (10, eRAPID; 6, UC). In the eRAPID arm, completion rates were higher in patients treated with prostateRT compared with chemoRT (week 1, 93% vs 69%; week 2, 93% vs 68%; week 12, 69% vs 55%). Overall, over 50% of online reports triggered self-management advice for milder adverse events. Unscheduled hospital contact was low, with no difference between eRAPID and UC. Return rates for outcome measures were excellent in prostateRT (97%-91%; 6-24 weeks) but lower in chemoRT (95%-55%; 6-24 weeks). Missing data were low (1%-4.1%), ceiling effects were evident in EQ5D-5L, self-efficacy-scale, and FACT-Physical Wellbeing. At 6 weeks, the chemoRT-eRAPID group showed less deterioration in FACT-G, EORTC QLQ-C30, and EQ5D-Visual Analogue Scale than UC, after baseline adjustment. CONCLUSIONS: eRAPID was successfully added to UC at 2 cancer centers in different patient populations. Acceptability and feasibility were confirmed with excellent adherence by prostate patients, but lower by those undergoing chemoRT for gynecological cancers.

Comparison of EORTC QLQ-C30 and PRO-CTCAE™ Questionnaires on Six Symptom Items.

CONTEXT: Clinical studies have over the past decade paid increasing attention to health-related quality of life data. Multiple questionnaires are often administered resulting in overlapping questions increasing patient burden. OBJECTIVES: To examine the correlations between the commonly used European Organization for Research and Treatment of Cancer Quality of Life Questionnnaire-C30 (QLQ-C30) and the Patient-Reported Outcomes version of the Common Terminology Criteria for Adverse Events (PRO-CTCAE™) on six coinciding items to determine consistency between overlapping items. METHODS: Data were prospectively collected from patients attending two cancer centers in the U.K. Participants completed the QLQ-C30 version 3.0 every four weeks and the PRO-CTCAE at least once a week for 12 weeks. Data were collected via the Internet or an interactive voice response. For the six coinciding items in QLQ-C30 and PRO-CTCAE: pain, nausea, vomiting, constipation, diarrhea, and fatigue, comparisons were made between all possible related responses by aligning the four responses in the QLQ-C30 with two condensed versions of the five responses in the PRO-CTCAE. Consistency and reliability was determined with the intraclass correlation coefficient (ICC) and Cronbach's α. RESULTS: About 247 patients completed 785 QLQ-C30 and 2501 PRO-CTCAE questionnaires. Moderate (ICC >0.5) to good (ICC >0.75) reliability and Cronbach's α >0.7 were found on all coinciding questions except for questions concerning the severity of nausea and vomiting as a result of relatively few patients responding to these questions. Items on frequency showed better correlations than the severity and interference items. CONCLUSION: The good reliability and consistency between the QLQ-C30 and PRO-CTCAE support future attempts to minimize patient burden by shortening health-related quality of life questionnaires.

Oncology professionals' views on the use of antidepressants in cancer patients: a qualitative interview study.

OBJECTIVES: Emotional distress, including depression, is an important issue for cancer patients and their families. Guidelines recommend the use of antidepressant drugs (ADs) for the management of depression in cancer. This study explores the views of oncology professionals about the inclusion of ADs in treatment plans. DESIGN: Semi-structured interview study. Data were analysed using framework analysis. SETTING: A specialist cancer centre and six district general hospitals across the Yorkshire Cancer Network. PARTICIPANTS: 18 randomly selected professionals from lung, breast, urology and colorectal cancer teams: oncologists (n=8), surgeons (n=3), clinical nurse specialists (n=2) and ward nurses (n=5). RESULTS: Three main themes emerged relating to professionals' attitudes, knowledge and behaviour. Positive attitudes were primarily expressed by nurses. However, negative views were expressed about the potential for over-reliance on ADs, and their use constituting 'giving in'. Doctors reported a lack of confidence in the use of and knowledge about ADs with an associated reluctance to prescribe. The general practitioner (GP) was regarded as the most appropriate professional to prescribe ADs. CONCLUSIONS: Cancer professionals highlighted a need for training in the appropriate use of ADs. Further, this research suggests that negative attitudes towards antidepressants may be a factor in their exclusion from treatment plans. The GP is seen to have a key prescribing role for AD therapy; however, it is unclear whether the GPs is asked to do this. This research raises questions about the adequacy of ADs in cancer care and to what extent the GP is able to meet this need.