Lung cancer researcher Solange Peters, MD, PhD, discusses the intersection of COVID-19 and cancer, as well as her research on immunotherapy biomarkers.


Solange Peters, MD, PhD [ESMO]

As of September 15, more than 930,000 deaths worldwide have been attributed to COVID-19, according to the Johns Hopkins dashboard, although the true number may be significantly higher. People with cancer, particularly lung cancer, may have the highest risk of dying from the infection, noted Solange Peters, MD, PhD, of the Centre Hospitalier Universitaire Vaudois in Lausanne, Switzerland, who studies the disease and helped create COVID-19 patient registries such as the COVID-19 and Cancer Consortium (CCC19) and the Thoracic cancERs international coVid 19 cOLlaboraTion (TERAVOLT). She is also president of the European Society for Medical Oncology (ESMO), which will virtually hold its annual congress on September 19–21. Peters shared her thoughts on the pandemic and insights on her lung cancer biomarker research with Cancer Discovery's Catherine Caruso.

With these registries, we wanted to answer some basic questions: Are cancer patients more vulnerable to COVID-19? Do they have a higher risk of complications and death? The main conclusion is that cancer patients are probably at higher risk than other patients. The most important factor in patients with cancer is whether they are consistently suffering from active disease. An even bigger factor might be if this disease is progressing or requires heavy treatment, including chemotherapy or steroids. Therefore, the most vulnerable patients have advanced disease that is not controlled, and within this category, the worst of the worst is lung cancer. This is probably because lung cancer is a more aggressive disease that causes impairment of lung function, which can be compounded by COVID-19. Mortality in cancer patients with COVID-19 ranges from 16% to 20% in CCC19 and 30% to 40% in TERAVOLT [compared with 3% to 4% generally]. If we face a second wave of COVID-19, we have to keep these vulnerabilities in mind, and this will guide us in protecting patients with cancer from exposure to the virus.

Also, in TERAVOLT, lung cancer patients had a mortality of 30% to 40%, but their rate of intensive care unit [ICU] admission was extremely low, 5% to 10% at the first analyses. Why are so many patients with lung cancer and COVID-19 dying but not being admitted to the ICU? It is probably due to health care systems being overwhelmed. It may also be due to local management rules where cancer patients with COVID-19 were not eligible for ICUs. A final barrier is that prior to being infected with COVID-19, patients with lung cancer may have discussed with their physicians a preference to not go to the ICU. Knowing this, we need to consider how we will handle lung cancer patients as we prepare for a second wave of COVID-19.

At the beginning of the year, the WHO [World Health Organization] and the European Parliament emphasized the need to reduce mortality from noncommunicable diseases or NCDs—and cancer certainly represents a big part of that. But then COVID-19 came, and the first thing that happened was we partially closed NCD facilities like cancer centers. Now screening rates for breast, colorectal, and lung cancers have dramatically gone down. This will translate not only into missed cancers, but also later-stage cancers—so higher mortality for sure. Everything about treatment delivery has been slowed down, meaning that some patients who should have received treatment did not. This will also translate into higher mortality. What will be the excess in cancer mortality 5 years from now? I fear it will be a substantial number, largely due to the COVID-19 pandemic, but not necessarily due to COVID-19 infections.

We are investigating how to optimize care for patients according to tumor biomarkers, as well as host biomarkers when it comes to immunotherapy. We started studying tumor-defining oncogenes—oncogene addiction—and we did a lot of trials around EGFR and, later on, ALK. I worked with international colleagues on the development of ALK inhibitors, and this year at ESMO, we will hear about the last compounds being evaluated for frontline use, such as lorlatinib [Lorbrena; Pfizer]. The one we studied the most, alectinib [Alecensa; Roche], has allowed patients to survive for years despite metastatic disease, which is absolutely amazing. When I started in lung cancer, the expected survival of metastatic patients was much less than a year.

Immunotherapy has completely changed the way we treat patients, with expectations that we never had 10 years ago, including long-term control of metastatic diseases. The next step is to increase this proportion of patients who experience a long-term benefit from immunotherapy, and this is about biomarkers.

With the Ludwig Institute for Immunotherapy of Cancer, we've developed a platform for biomarker evaluation, so we are the central lab of many pharma-driven trials. We are investigating questions like these: Who are the patients that benefit from pembrolizumab [Keytruda; Merck] in the adjuvant setting? Who are the patients that benefit from frontline ipilimumab [Yervoy; Bristol-Myers Squibb (BMS)] and nivolumab [Opdivo; BMS] in mesothelioma? In the oncology department we have a set of tools and almost 50 labs looking at immunotherapy biomarkers. It's really a machine to find the best predictors so we can choose the right treatment for the patient.

We are also starting BIOMARK, a real-world trial for patients on immunotherapy. We are going to collect urine, blood, plasma, serum, stool, and tumor biopsies, and analyze everything in parallel to describe the context of an immune response, as well as the risk of toxicities. We'll look at how the immune system might be influenced by the microbiome, look at the neoantigens in the tumor and the corresponding reactive T cells, as well as the modulation that can happen in the immune microenvironment of the tumor, and finally the cytokine milieu. We will evaluate these biomarkers in all patients. We are at a point where this information helps, but we still don't know how to derive a biomarker signature to predict who should receive the different immunotherapies. We need to identify the patients who will benefit from each drug, and the patients who will not respond.