Palliative Care

PV QA 3 - Poster Viewing Q&A 3

TU_31_3043 - Survival for patients with metastatic disease - How good are radiation oncologists at predicting survival?

Tuesday, October 23
1:00 PM - 2:30 PM
Location: Innovation Hub, Exhibit Hall 3

Survival for patients with metastatic disease – How good are radiation oncologists at predicting survival?
D. C. Ling1, S. Beriwal2, D. L. Beichner1, and D. E. Heron2; 1Department of Radiation Oncology, UPMC Hillman Cancer Center, University of Pittsburgh School of Medicine, Pittsburgh, PA, 2UPMC Hillman Cancer Center, Pittsburgh, PA

Purpose/Objective(s): Patients and physicians alike seek an accurate estimate of survival following a metastatic cancer diagnosis, yet reported median survival outcomes may not reliably account for the unique circumstances of individual patients. At our institution, as part of the pathway entry for all patients with non-curative treatments, the physician enters whether he or she would be surprised if the patient dies within the next year. We thus aimed to determine the accuracy of physicians’ predictions of patient survival within a large network.

Materials/Methods: We performed a retrospective review of 1586 patients across a network, consisting of 35 physicians at 18 sites, who were treated with palliative-intent radiation therapy for metastatic, recurrent, or primary non-recurrent disease between February 2016 and December 2016. All patients evaluated after December 31, 2016 were excluded in order to allow for at least 1 year of follow-up. For each patient’s first visit with a given physician, the physician’s pathway entry with regard to expected survival was reviewed. A chi-square analysis was performed to determine the accuracy of physicians’ predictions as to whether the patient would survive for 1 year following the first appointment date. Univariate and multivariate binary logistic regression were performed to identify any potential predictors of the physicians’ accuracy, such as academic vs. community setting, physician’s number of years in practice, and disease site.

Results: Among 843 patients who were expected by their physicians to live past 1 year, 455 (54.0%) died within 1 year, and of 743 patients who were not expected to live past 1 year, 602 (81.0%) died within 1 year (p<0.001). On univariate analysis, there was no difference in the accuracy of survival predictions according to academic vs. community practice setting (63.8% vs. 60.9% of predictions accurate, respectively) (p=0.236), but physicians in practice for <15 years had higher accuracy than physicians in practice for ≥15 years (65.4% vs. 60.5% of predictions accurate, respectively) (p=0.047), and predicted prognosis was accurate less often for patients treated for brain metastases (55.6%) compared to those treated for bone metastases (64.6%) or other disease sites (68.7%) (p<0.001). On multivariate analysis, only disease site remained significant (p<0.001). Among patients who died within 1 year, the median time to death was 2.4 (range: 0.03-11.90) months for those not expected to live past 1 year, and 4.1 (range: 0.13-11.93) for those expected to live past 1 year.

Conclusion: Physicians tend to overestimate survival for patients treated with palliative-intent radiation therapy for metastatic, recurrent, or primary non-recurrent disease. Survival predictions are more accurate for patients whom physicians do not expect to survive past 1 year, than for those who are expected to survive past 1 year. Better objective tools are needed to aid physicians in predicting survival so that patients and families can be guided appropriately.

Author Disclosure: D.C. Ling: None. S. Beriwal: None. D.L. Beichner: None. D.E. Heron: No personal compensation; Accuray Exchange in Radiation Oncology. Partnership; Cancer Treatment Services International. Vice Chairman of Clinical Affairs; University of Pittsburgh School of Medicine. Director of Radiation Services; UPMC CancerCenter.

Diane Ling, MD

University of Pittsburgh Cancer Institute

Presentation(s):

Send Email for Diane Ling


Assets

TU_31_3043 - Survival for patients with metastatic disease - How good are radiation oncologists at predicting survival?



Attendees who have favorited this

Please enter your access key

The asset you are trying to access is locked. Please enter your access key to unlock.

Send Email for Survival for patients with metastatic disease - How good are radiation oncologists at predicting survival?