ICU Admission, Oxygen Use Predict Outcomes in Patients With Cancer and COVID-19

Markus Joerger, MD, discussed the outcomes and prognostic factors of patients with cancer won contract COVID-19 in an interview with Targeted Oncology.

It is well known that patients with cancer who are infected with COVID-19 have worse outcomes than individuals without cancer. However, questions remain around specific prognostic factors.

An analysis of 455 Swiss patients looked at the outcomes and prognostic factors of patients with cancer who contract COVID-19. Cancer types represented in the analysis included breast (14%), lung (10%), prostate (6%), and myeloma (4%).

Nearly half, 45%, had non-curative disease and 23% of patients received systemic treatment within 3 months prior to COVID-19 diagnosis. In total, 63% were hospitalized for COVID-19, 47% required oxygen, and 9% required invasive ventilation.

Lead author Markus Joerger, MD, of the Medical Oncology and Hematology Department at Kantonsspital St. Gallen in St. Gallen, Switzerland, discussed the outcomes and prognostic factors of patients with cancer won contract COVID-19 in an interview with Targeted OncologyTM.

TARGETED ONCOLOGY™: Can you give an overview research on the outcomes and prognostic factors of COVID-19 infection in cancer patients?

JOERGER: At the start of the pandemic, we realized that it would be important to have some data on the outcome of cancer patients [infected with COVID-19], because obviously, cancer patients are a group of people that are more vulnerable to COVID-19 infection. And so, what we did is we collected data from all cancer patients that had a symptomatic COVID infection over a 1-year period starting March 1, 2020. The majority of patients had a polymerase chain reaction confirmed COVID infection. All were symptomatic. And we just looked at prognostic factors.

Can you go into some of the results that you observed?

We had 455 patients. A slight majority was 65 or older. So, an elderly population, probably older than a standard population, if you do not look specifically at cancer. Then solid tumors versus hematologic tumors, that was the majority, 75% were solid tumors. And then thirdly, 50% were in a non-curative setting. And we also looked at previous treatment before the diagnosis of COVID within the 3-month time period before the diagnosis. A fifth of the patients received chemo, a fifth received targeted agents, and a minority of 7% received checkpoint inhibitors. Then we looked at comorbidity. Many patients had cardiovascular comorbidity, more than 50%. And close to 15%, had lung disease comorbidities.

This is just the characteristics of the population, but what is certainly more important is the outcome of the COVID infection. And these results were pretty dark so to say. More than 60%, 62% to be exact, were hospitalized for COVID. There is a footnote there in that some patients were already hospitalized, but probably all patients were hospitalized for the symptomatic COVID infection. Almost 50%, 47% to be exact, required oxygen. And, importantly, almost 40%, 34.6% to be exact, were admitted to the intensive care unit (ICU) and almost 10% required invasive ventilation. So pretty severe consequences.

And then of course, most importantly, COVID-19 mortality in all patients was 21.5%. In hospitalized patients, it was 31%. In patients that required oxygen, 41%, and in ICU patients, 56%. The overall mortality of more than 20% is clearly higher than what we know from the general population. Of course, this is not a clean comparison.

We looked a little bit at the specific treatment, but I think that is probably not that important. What is certainly important is that we did univariate and multivariate analysis to look at significant prognostic factors for the COVID-19 outcome. And here, there were basically no surprises. Oxygen requirement was a very heavy negative prognostic factor with a hazard ratio of 22. ICU admission, of course, also with a hazard ratio of 4.4. This is multivariate now, age 65 or older with a hazard ratio of 3.2. And the noncurative versus curative setting with a with a hazard ratio of 2.4. So, these were the significant multivariant negative prognostic factors. So, no surprises there. And in conclusion, on one hand, the severe outcome of COVID-19 in this patient population. It pretty comparable to other countries, though. That also interesting for us, because, in Switzerland, you have a centralized health care system. That somehow different than if you compare it to the UK, or to the US or many other countries. So, it was also interesting to understand whether a decentralized system would fare better or worse in the pandemic with regards to cancer patients. And that seems not to be the case.

Did patients with certain cancer types fair better than others?

We looked at that, and there are other studies that suggested that hematologic malignancies had somewhat worse outcomes, but we didn't see that now. And the numbers for very specific tumor types were too small to really have reliable insight into that. So, we did this analysis, but we didn't see that particular cancer types did particularly well or particularly bad.

Do you have any research planned for the future?

We are not directly using this data set, but upcoming data sets. And this topic is the success of vaccinations in cancer patients. There is already some insight into that from other countries, but also from our group and in our country that vaccines really do not work as good in cancer patients as in healthy, healthy people. And this is a big concern, and it is, of course, very important to tackle this issue.


Joerger M, Metaxas Y, Zaman K, et al. Outcome and prognostic factors of COVID-19 infection in cancer patients: Final results of SAKK 80/20. Ann Oncol. 2021;32(5):1134-1135. doi: 10.1016/j.annonc.2021.08.1563