Warner JL, et al. Abstract LBA110. Presented at: ASCO20 Virtual Scientific Program; May 29-31, 2020.
Read Next
Progressing malignancy linked to higher mortality among patients with cancer and COVID-19
Warner JL, et al. Abstract LBA110. Presented at: ASCO20 Virtual Scientific Program; May 29-31, 2020.
Click Here to Manage Email Alerts
Progressing cancer appeared independently associated with increased risk for death among a cohort of patients with malignancies and COVID-19, according to study results presented during the ASCO20 Virtual Scientific Program.
Other factors linked to higher 30-day mortality among this cohort included age, male sex, ECOG performance status.
Receipt of a combination of hydroxychloroquine and azithromycin also appeared associated with higher mortality risk; however, researchers emphasized no conclusions can be drawn about the impact of this combination because several factors — including disease severity — may have influenced which patients received it, therefore confounding the results.
“The effects of COVID-19 on patients with cancer remain poorly understood,” Jeremy Lyle Warner, MD, MS, medical director of the Vanderbilt Cancer Registry and associate professor of medicine and biomedical informatics at Vanderbilt-Ingram Cancer Center, said during a presentation. “There are some published reports, but they tend to be on the smaller size or are focused on a specific geographic region. We wanted to collect a larger and more representative cohort to learn more about COVID-19 and cancer.”
Warner presented results based on cases reported by academic and community hospitals in the COVID-19 and Cancer Consortium, which is collecting data on patients with active or previous hematologic or invasive solid malignancies.
Warner and colleagues analyzed 928 patients (median age, 66 years; range, 18-90) with cancer and COVID-19 to determine who may be most at risk for mortality and to identify potential prognostic factors.
The most common malignancies in the cohort included breast cancer (20%) and prostate cancer (16%), and 43% of the total population remained on active anticancer treatment.
About 20% of patients received the combination of hydroxychloroquine and azithromycin, and about 10% received hydroxychloroquine alone, Warner said.
At the time of data analysis, 106 patients had died and 26% met the composite outcome of death, severe illness requiring hospitalization or mechanical ventilation.
Within 30 days of COVID-19 diagnosis, 121 patients (13%) had died.
Multivariable logistic regression analysis identified several independent factors associated with higher 30-day mortality. These included age, male sex, former smoking, ECOG performance status (ECOG 2 vs. 0 or 1: partially adjusted OR [paOR] = 2.74; 95% CI, 1.31-5.7; ECOG 3 or 4 vs. ECOG 0 or 1: paOR = 5.34; 95% CI, 2.44-11.69), active malignancy status (stable/responding: paOR = 1.93; 95% CI, 1.06-3.5; progressing: paOR = 3.79; 95% CI, 1.78-8.08), and receipt of azithromycin in combination with hydroxychloroquine.
Tumor type, race, ethnicity, obesity, number of comorbidities, recent surgery and type of cancer therapy did not appear to be significant predictors of mortality.
“The patients who were hospitalized were much more likely to receive hydroxychloroquine or the combination of [hydroxychloroquine and azithromycin,” Warner said. “Our data are limited about when exactly these patients received the drugs and require follow-up. It’s too early to make any major conclusions on these data regarding the combination of these drugs.” – by John DeRosier
References:
Warner JL, et al. Abstract LBA110. Presented at: ASCO20 Virtual Scientific Program; May 29-31, 2020.
Disclosures: Warner reports consultant/advisory roles with and/or travel expenses from IBM and Westat and stock ownership in HemOnc.org. Please see the abstract for all other researchers’ relevant financial disclosures.
Perspective
Back to Top
Martin J. Edelman, MD, FACP
Warner and colleagues presented data from nearly 1,000 people with cancer who were infected with COVID-19. Similar to the initial report from China and other studies from the United States and Europe, patients with cancer appeared to be at higher risk for mortality than the general population. Similar to reports for the general population, older individuals and those with obesity and smoking history do worse. Not surprisingly, patients with debility (ie, poor performance status) and progressive malignancy also do worse. Although the abstract states there was no interaction with specific diseases or treatments, it is unlikely that the study would have had adequate power to resolve these question.
It is interesting that there was higher mortality associated with the use of azithromycin and hydroxychloroquine. This increased mortality could be the result of direct toxicity or the use of the agents by patients who were sicker. This again demonstrates the need for controlled trials of therapeutics and caution in terms of adopting anecdotal results for patient care. This pandemic once again demonstrates the importance of carefully considered and properly performed clinical research.
Perspective
Back to Top
Lee Greenberger, PhD
This was an expected outcome because there have been a plethora of other papers already published that have shown similar results. Patients with cancer who have to enter the hospital tend to already have other comorbidities and may be sicker than other patients so, when you add COVID-19 on top of that, it’s going to affect potential outcomes.
You would assume that people with lung cancer would be at one of the greatest risks, however, we haven’t seen strong correlation data in the literature yet between types of cancer and risks for death related to COVID-19.
This study showed that some patients took hydroxychloroquine, sometimes in combination with azithromycin. Based on this abstract, we can’t make any conclusion about these drugs. However, we can probably make conclusions on hydroxychloroquine based on a number of other studies that have been published. They have shown that the drug, whether taken alone or in a combination, is probably not effective as a therapeutic. What we don’t know is how sick some of these patients were and whether they were on ventilators or whether they were taking it as a preventive measure. This particular abstract, I think, raises more questions than it answers in terms of hydroxychloroquine.
Click Here to Manage Email Alerts
