Unnecessary Use of Radiotherapy Persists in End-of-life Patients

News

The use of unnecessary radiotherapy in patients with metastatic cancer who are near the end of life is still far higher than guidelines suggest it should be, concludes a nationally comprehensive database study. It underscores the need for better adherence to the guidelines, say the authors.

The study was published online January 14 in JAMA Health Forum.

“No patient should be subjected to extended or protracted treatments that are not evidence-based,” lead author Patricia Santos, MD, from the Department of Radiation Oncology at Memorial Sloan Kettering Cancer Center, New York City, told Medscape Medical News in an email.

“Given that patients at the end of life are a particularly vulnerable population, it should concern practitioners that extended radiotherapy is still quite common in this population,” she added.

Senior author Erin Gillespie, MD, also from Sloan Kettering, agreed. “Our findings should matter to patients and practitioners alike as overtreatment or unnecessary treatment of any kind can increase the risk of potential harm without proven benefit.”

“And stakes are arguably highest for patients with metastatic cancer, for whom quality not quantity of life is our primary goal,” Gillespie emphasized. The study results underscore “the importance of not offering extended treatments to any patient, as some will invariably end up spending a disproportionate amount of their remaining days on treatment,” she said.

Guideline-Nonconcordant Radiotherapy

The analysis was limited to collecting data on patients who were age 65 years or older and who died within 90 days of treatment planning.

The primary outcome was guideline-nonconcordant radiotherapy, the authors note. Santos explained that radiotherapy was considered to be ‘nonconcordant’ if the number of treatments exceeded 10 fractions or the radiation was delivered using a technique that has not been recommended by the American Society of Radiation Oncology.

As part of the Choosing Wisely campaign, ASTRO has broadly stated that practitioners should not routinely use extended fractionation schemes for palliation of bone metastases, nor should they routinely add adjuvant whole brain radiation therapy to stereotactic radiosurgery for limited brain metastases.

For the study, the team analyzed 467,781 radiotherapy episodes.

Of these, 16% were using radiotherapy to target bone and brain metastases (9.2% were bone metastases, 5.8% were brain metastases).

Some 3.7% of the group died within 90 days of radiotherapy, roughly half of whom died of bone metastases and the other half of brain metastases.

Of the patients who died within 90 days of receiving radiotherapy, 78.4% received guideline-concordant radiotherapy but the other 21.6% received guideline-nonconcordant radiotherapy, the authors report.

On multivariate analysis, patients who were treated in a hospital-affiliated facility were half as likely to receive guideline-nonconcordant radiotherapy at an adjusted odds ratio (aOR) of 0.50.

Older patients between age 75-85 years were 10% less likely to receive guideline-nonconcordant radiotherapy at an aOR of 0.90, they add, while patients 85 years of age and older were 27% less likely to receive guideline-nonconcordant radiotherapy at an aOR of 0.73. Both age groups were being compared to patients between 65 and 75 years of age.

Conversely, patients who had undergone a major procedure were 17% more likely to receive guideline-nonconcordant radiotherapy, while those who had received chemotherapy were 26% more like to receive the same inappropriate treatment.

Interestingly, patients who were expected to survive for more than 30 days after the treatment planning appointment were at greatest risk to receive guideline-nonconcordant therapy, the investigators point out.

For example, patients whose survival was projected to last for up to 60 days after the treatment planning appointment were almost five times as likely to receive guideline-nonconcordant radiotherapy at an aOR of 4.72. Those  who were projected to survive up to 90 days after the treatment planning appointment were almost 7 times more likely to receive guideline-nonconcordant radiotherapy at an aOR of 6.55. Both groups were being compared with patients whose survival was not projected to go beyond 30 days (P <.001 for both endpoints).

“In general, our results showed that the odds of receiving nonguideline concordant radiation were higher if the patient lived longer than 30 days after treatment,” Santos acknowledged. “This probably speaks to the fact that patients who are perceived to have a better prognosis are often given more aggressive forms of care,” she added.

However, the fallacy in that thinking, Santos pointed out, is that prognosis can be very difficult to assess in patients with metastatic disease, especially when the cancer has moved into the bone and brain.

Although the claims data analyzed in this study did not allow investigators to assess performance status at the time of treatment, “our guiding principle was that the guidelines generally apply to all patients regardless of prognosis,” she emphasized.

Furthermore, the fact remains, Santos noted, that the goal of care for any individual patient can be dynamic at least on treatment initiation; if that goal moves toward improving survival, “treating a patient with definitive vs palliative intent can drastically change the ways in which radiation is delivered.”

This problem might be remedied by involving palliative or supportive care services earlier in a patient’s treatment journey as recruitment of these services early may minimize the risk of overtreatment, Santos suggested.

Gillespie reports funding from the National Cancer Care Network/Pfizer EMBRACE, and was a co-founder of an educational website for radiation oncology professionals (eContour.org). No other disclosures were reported.

JAMA Health Forum. Published online January 14, 2022. Abstract

For more from Medscape Oncology, join us on Twitter and Facebook

Leave a Reply

Your email address will not be published. Required fields are marked *