Prognostic Tool Accurately Predicts Cancer Survival Risk

A prognostic survival model published in JAMA Network Open was able to accurately predict changing cancer survival risk over time using clinical, symptom, and performance status data.¹

Moreover, the model, titled PROVIEW, appears to have the potential to be a useful prognostic tool that can be completed by patients. The researchers also suggested that these findings may support the earlier integration of palliative care. 

“Because the model covariates can be completed by patients, PROVIEW may be a useful patient- facing online tool, allowing them to prepare questions around goals of care and treatment preferences before an oncologist visit,” the authors wrote. “In this way, PROVIEW could help patients and families initiate conversations with practitioners about the changing disease trajectory and explore the benefits of palliative care supports earlier.”

In this retrospective, population-based, prognostic study, researchers used data of patients diagnosed with cancer from January 1, 2008 to December 31, 2015 in Ontario, Canada. Using the Symptom Management database, which began in 2007 when Cancer Care Ontario mandated the systematic screening of outpatients with cancer for symptoms using the Edmonton Symptom Assessment System (ESAS) and for performance status using the Palliative Performance Scale (PPS), researchers were able to collect validated data on the patients.  

The ESAS asks patients to self-report the severity of 9 symptoms, including pain, depression, well-being, shortness of breath, anxiety, nausea, tiredness, drowsiness, and appetite, on a scale of 0 (symptom absent) to 10 (most severe), whereas the PPS describes a patient’s performance status based on a patient’s level of ambulation, level of activity, and ability to perform self-care. In 2013, Ontario also began collecting functional scores using a patient-completed Eastern 

Cooperative Oncology Group score, which is comparable to the physician-reported PPS. 

Patients were randomly selected for model derivation (60%) and validation (40%). The derivation cohort was then used to create a multivariable Cox proportional hazards regression model with baseline characteristics under a backward stepwise variable selection process to predict the risk of mortality as a function of time.

Overall, 255,494 patients diagnosed with cancer were identified for the study. Researchers recalculated time to death from diagnosis (year 0) at each of 4 annual survivor marks after diagnosis (up to year 4), and the cohort decreased to 217,055, 184,822, 143,649, and 109,569 patients for each of the 4 years after diagnosis.

In the derivation cohort at year 0, the most common cancer types were breast (30,855 [20.1%]), lung (19,111 [12.5%]), and prostate (18,404 [12.0%]). Moreover, a total of 47,614 (31.1%) had stage III or IV disease and the mean (SD) time to death in year 0 was 567 (715) days.

After backward stepwise selection in year 0, the following factors were correlated with increased risk of death by more than 10%: 

• Being hospitalized

• Having congestive heart failure

• Chronic obstructive pulmonary disease, or dementia

• Having moderate to high pain

• Having worse well-being

• Having functional status in the transitional or end-of-life phase

• Having any problems with appetite

• Receiving end-of-life home care

• Living in a nursing home

Further, Model discrimination was found to be high for all models (C statistic: 0.902 [year 0], 0.912 [year 1], 0.912 [year 2], 0.909 [year 3], and 0.908 [year 4]). 

“To our knowledge, PROVIEW is the only cancer prognostic model that uses these patient-reported outcomes and updates the risk yearly after diagnosis. Because the covariates are self-reportable by patients and predict risk in days, the model has potential to be a patient-completed online tool, allowing patients to examine survival predictions during various periods as their condition changes.” the authors wrote. “Patients can use the model’s survival predictions, which uniquely incorporate changes in symptoms, performance status, treatment, and hospital use along the disease trajectory, to inform discussions and improve decision-making with practitioners.”

In an invited commentary, Laura Van Metre Baum, MD and Debra Friedman, MD, both of the Vanderbilt University Medical Center, indicated that as oncology care continues to evolve, early and accurate predictors of mortality and morbidity will become even more important.² However, they suggested that though the PROVIEW model is comprehensive, the widespread implementation and dissemination of the tool will require further study. 

“Although the numeric probability of survival in days may be of interest to some patients, a discussion with a practitioner of what this means-in hours to days, days to weeks, weeks to months, or months to years-will be needed to assist patients in appropriate legacy planning, EOL preparedness, and medical decision-making,” the authors wrote in the commentary. “Fundamentally, the model provides information, with the eventual goal to empower patients to speak with practitioners about palliative care. The translation of knowledge to behavior and of information to empowerment and activation requires further intervention for successful implementation.”

References:

1. Seow H, Tanuseputro P, Barbera L, et al. Development and Validation of a Prognostic Survival Model With Patient-Reported Outcomes for Patients With Cancer. JAMA Network Open. doi:10.1001/jamanetworkopen.2020.1768.

2. Baum LVM, Friedman D. The Uncertain Science of Predicting Death. JAMA Network Open. doi:10.1001/jamanetworkopen.2020.1736.