The findings have prompted an expert to argue against the use of immunotherapy for such patients, who may have little time left and very little chance of benefiting.
“It is quite clear from clinical practice that most patients with limited PS do very poorly and do not benefit from immune check point inhibitors (ICI),” Jason Luke, MD, UPMC Hillman Cancer Center and the University of Pittsburgh, said in an email.
“So, my strong opinion is that patients should not be getting an immunotherapy just because it might not cause as many side effects as chemotherapy,” he added.
“Instead of giving an immunotherapy with little chance of success, patients and families deserve to have a direct conversation about what realistic expectations [might be] and how we as the oncology community can support them to achieve whatever their personal goals are in the time that they have left,” he emphasized.
Dr. Luke was the lead author of an editorial in which he commented on the study. Both the study and the editorial were published online in JCO Oncology Practice.
Variety of cancers
The study was conducted by Mridula Krishnan, MD, Nebraska Medicine Fred and Pamela Buffett Cancer Center, Omaha, Nebraska, and colleagues.
The team reviewed 257 patients who had been treated with either a programmed cell death protein–1 inhibitor or programmed cell death–ligand-1 inhibitor for a variety of advanced cancers. The drugs included pembrolizumab (Keytruda), nivolumab (Opdivo), atezolizumab (Tecentique), durvalumab (Imfinzi), and avelumab (Bavencio).
Most of the patients (71%) had good PS, with an Eastern Cooperative Oncology Group (ECOG) PS of 0-1 on initiation of immunotherapy; 29% of patients had poor PS, with an ECOG PS of greater than or equal to 2.
“The primary outcome was OS stratified by ECOG PS 0-1 versus ≥2,” note the authors. Across all tumor types, OS was superior for patients in the ECOG 0-1 PS group, the investigators note. The median OS was 12.6 months, compared with only 3.1 months for patients in the ECOG greater than or equal to 2 group (P < .001).
Moreover, overall response rates for patients with a poor PS were low. Only 8%, or 6 of 75 patients with an ECOG PS of greater than or equal to 2, achieved an objective response by RECIST criteria.
This compared to an overall response rate of 23% for patients with an ECOG PS of 0-1, the investigators note (P = .005).
Interestingly, the hospice referral rate for patients with a poor PS (67%) was similar to that of patients with a PS of 1-2 (61.9%), Dr. Krishnan and colleagues observe.
Those with a poor PS were more like to die in-hospital (28.6%) than were patients with a good PS (15.1%; P = .035). The authors point out that it is well known that outcomes with chemotherapy are worse among patients who experience a decline in functional reserve, owing to increased susceptibility to toxicity and complications.
“Regardless of age, patients with ECOG PS >2 usually have poor tolerability to chemotherapy, and this correlates with worse survival outcome,” they emphasize. There is as yet no clear guidance regarding the impact of PS on ICI treatment response, although “there should be,” Dr. Luke believes.
“In a patient with declining performance status, especially ECOG PS 3-4 but potentially 2 as well, there is little likelihood that the functional and immune reserve of the patient will be adequate to mount a robust antitumor response,” he elaborated.
“It’s not impossible, but trying for it should not come at the expense of engaging about end-of-life care and maximizing the palliative opportunities that many only have a short window of time in which to pursue,” he added.
Again, Dr. Luke strongly believes that just giving an ICI without engaging in a frank conversation with the patient and their families – which happens all too often, he feels – is absolutely not the way to go when treating patients with a poor PS and little time left.
“Patients and families might be better served by having a more direct and frank conversation about what the likelihood [is] that ICI therapy will actually do,” Dr. Luke stressed.
In their editorial, Dr. Luke and colleagues write: “Overall, we as an oncology community need to improve our communication with patients regarding goals of care and end-of-life considerations as opposed to reflexive treatment initiation,” he writes.
“Our duty, first and foremost, should focus on the person sitting in front of us – taking a step back may be the best way to move forward with compassionate care,” they add.
The authors and editorialists have disclosed no relevant financial relationships.
A version of this article first appeared on.