Negotiating the Obstacles to Conducting Clinical Trials of Immunotherapy During the Coronavirus Pandemic

A Conversation With Charles G. Drake, MD, PhD

Get Permission

Coronavirus disease 2019 (COVID-19) is perhaps the biggest challenge health-care systems have ever had to face. As part of a series of interviews The ASCO Post is conducting with oncologists, we talked with Charles G. Drake, MD, PhD, about the impact of COVID-19 on his practice and on the conduct of clinical trials of immunotherapy in patients with genitourinary cancers. Dr. Drake leads genitourinary oncology and is Co-Director of the Cancer Immunotherapy Programs at Columbia University’s Herbert Irving Comprehensive Cancer Center, New York City—the epicenter of the COVID-19 outbreak.

Charles G. Drake, MD, PhD

Charles G. Drake, MD, PhD

This interview reflects conditions as of April 3, 2020.

Protection for Physicians

How are you and your colleagues protecting yourselves from COVID-19 infection?

For Columbia inpatients, we are already used to frequently donning gowns and gloves for patients with vancomycin-resistant enterococcus. Since COVID-19 is more infectious, we use additional protective equipment, particularly a well-fitted N95 mask, a disposable mask over that, and gloves. For eye protection, we use goggles or face shields. This system preserves personal protective equipment, as both the goggles and N95 masks are reused. Maybe it’s a little excessive, but I go home, take off my scrubs, and put them immediately into the washing machine and then take a long, hot shower. Whenever I come in from anywhere,
I wash my hands thoroughly.

Coverage by Inpatient Cancer Service

Are you seeing patients with COVID-19?

Our hospital was filling up quickly, and the inpatient cancer service was asked to take care of a number of medical patients. About two-thirds of the patients we are covering are general medicine patients, and most have tested positive for COVID-19. Of course, patients with serious respiratory problems who require a ventilator are moved to the intensive care service.

It is possible to keep a patient with breathing difficulty out of the intensive unit by using a nonrebreathing mask or high-flow nasal cannula. A number of patients have been kept on the high-flow cannula before they go onto a ventilator or die; some patients do come off a ventilator and are able to go home.

COVID-19 and Clinical Trials at Columbia

How has COVID-19 affected the conduct of clinical trials at Columbia?

At first, we realized the hospital would fill up with patients who had COVID-19, and we would have to close most of our clinical trials. We had to decide which trials would be closed for enrollment. We decided that patients already on a clinical trial could stay on that trial.

At that time, we decided to keep two types of trials open: one was for patients who had other treatment options but who we thought could benefit from immunotherapy. For example, patients with prostate cancer whose disease progressed on hormonal therapy and chemotherapy were enrolled in a trial of a bispecific antibody. At that time, the operating room (OR) was reserved for essential surgery and patients on clinical trials, so we also tried to keep our neoadjuvant trials open, where patients received treated before surgery.

One week later, we learned the ORs would be closed completely and not used for clinical trials. Although we were told to close enrollment in our neoadjuvant trials, we were able to get one patient to surgery just in time and obtained a tissue sample for the lab to evaluate whether the immunotherapy was working.

More recently, our end-stage trial for prostate cancer was closed to enrollment. So now, all our clinical trials are closed to new enrollment.

How many clinical trials of immunotherapy for genitourinary cancers are open?

Currently, we have three neoadjuvant trials for prostate, kidney, and bladder cancers, respectively. The hope is that immunotherapy will enable these patients to have surgery. We have one end-stage, life-saving trial of a bispecific antibody that uses prostate-specific membrane antigen to target T cells to prostate tumors through CD3.

Managing Patients Already on Clinical Trials

What are the current logistics for monitoring patients who are already on clinical trials?

For some patients enrolled on trials who are appropriate for routine follow-up, we now do these visits by telemedicine. For patients with cancer on clinical trials who need to come in for monitoring or treatment, our cancer center is open for treatment.

I’ve been doing telemedicine appointments for about 3 weeks now. The patient goes into a virtual waiting room, and we use Epic software. It’s not difficult to do if you know the patients well and how they communicate about themselves, but telemedicine is not as useful if you don’t already know the patient. With someone new, it is difficult to read body language and get a sense of performance status (robust or frail). And of course telemedicine doesn’t take the place of physical exam.

Some of our patients have to travel from out of state to our clinic but are now forbidden to travel because of the COVID-19 pandemic. In those cases, most clinical trial sponsors have relaxed their policies to allow receiving lab tests at laboratory facilities near patients’ homes instead of at Columbia.

To Travel or Not to Travel to New York

How do you decide whether a patient should travel to the Columbia or not?

It is challenging to set up monitoring for patients on clinical trials. We have the conversation every day with patients: “Is it worth it for you to travel into New York to be monitored?”

Ultimately, you weigh the risks vs the benefits. The current prevalence of COVID-19 in the community is around 20%, and patients in the United States probably have around a 1% chance of dying if infected. The chances of dying from cancer depend on the type of cancer and the stage. Some patients with prostate cancer who have experienced disease progression through all lines of therapy have a 50% to 75% chance of dying from their cancer in the next year. The risk/benefit ratio tells them to continue on a clinical trial.

However, the risk-benefit ratio is different for patients with prostate cancer enrolled in the MAGIC-8 trial. These men have earlier-stage disease and a rapid prostate-specific antigen doubling time, so the chance of their dying in the next year is low—no more than 10% to 15%.

It is important to reassure patients [on clinical trials] who are coming in for follow-up visits that they are going to be safe if they come in for clinic visits.

Screening Patients and Testing

What practices are in place to reassure patients about COVID-19?

Our cancer center has stepped up to the plate. At the Herbert Irving Cancer Center, a patient first encounters two or three nurses who are wearing gloves and goggles; they screen all who come into the building and take their temperature. If someone has any signs or symptoms suggesting COVID-19, that person is tested. Currently, it takes 4 to 6 hours to get test results, so, if they test negative for COVID-19, they are rescheduled for a visit over the next 1 or 2 days.

This screening process reassures patients there will be no one upstairs in the clinic who is infected. However, this is not a perfect system because asymptomatic patients can carry the virus. We are also not allowing visitors up to the treatment center; a similar policy is in place throughout the inpatient wards.

What about the false-negative results and other problems reported with COVID testing?

From what I understand, the polymerase chain reaction test itself is fairly accurate; most of the false-negative results are likely due to an inadequate sample. You have to swab deep, hard, and long into the nasopharyngeal cavity, which is uncomfortable to say the least. If one does a cursory swab, it is more likely to be a false-negative result.

Immunotherapy Trials at Columbia

Can you describe the immunotherapy trials you mentioned in greater detail?

The three neoadjuvant trials are for patients with high-risk localized cancer for whom surgery is not likely to be curative. In animal models, including metastatic models, immunotherapy before surgery is better than when given later. Giving neoadjuvant immunotherapy seems to increase the chances of curative surgery.

The neoadjuvant bladder cancer trial is randomized to compare 8 weeks of treatment with chemotherapy vs chemotherapy plus immunotherapy.

The neoadjuvant kidney cancer trial is a single-arm, investigator-initiated (Columbia alone) trial designed to evaluate anti–programmed cell death protein 1 (PD-1) plus anti–interleukin 1 beta. Preliminary results in animal models were exciting, but since COVID-19, no patients have enrolled in this trial. This is one of the saddest to me because the first couple of patients we treated on trial had rather astounding changes in the immune cells in their excised tumors.

The neoadjuvant prostate cancer trial evaluated an anti–cytotoxic T-lymphocyte–associated protein 4 nonfucosylated agent, similar to ipilimumab but designed to deplete regulatory T cells. We were able to enroll two patients on this trial before the COVID-19 crisis. Our idea for keeping these three trials open was to enable surgery for these patients—to provide them with some potentially meaningful treatment until the ORs open again.

Regulatory T cells are upregulated in prostate cancer and increase on hormonal therapy, according to research we have published.1 The idea is to give hormonal therapy to kill tumor cells and use this new agent to deplete the regulatory T cells.

The fourth trial is a single-arm, phase I trial of a bispecific antibody. We are excited about this because there is a suggestion of meaningful activity in patients with metastatic prostate cancer. We identified two patients who might enroll on this trial, before all trials were closed to enrollment, but just one of them was able to go on the trial. The other patient had lab values that precluded enrollment.

Immunotherapy and COVID-19

Have any of your patients on clinical trials of immunotherapy developed COVID-19?

None of mine have yet, but it is inevitable, as cases peak over the next week or two.

If these patients did develop COVID-19 infection, would they be sicker than other infected patients?

I have seen patients with the infection who are on checkpoint inhibitors, and they don’t seem to be more or less sick than other patients. I would have expected to see some difference, but, thus far, we haven’t, although it is still early. Mouse models of viruses and anti–PD-1 are inconclusive, although some suggest that PD-1 may facilitate viral clearance.

Protective Equipment and Ventilators

What is the situation at Columbia in terms of protective equipment?

From my limited perspective, so far, we have been fine at Columbia; we seem to have plenty of yellow gowns and purple exam gloves. Reusing N95 masks (when covered) and goggles seems to me to be a good solution. We are generally using the N95s for 3 to 5 days. The idea is to protect the types of masks painters wear with an inexpensive surgical mask and conserve the precious N95 masks. The duckbill and birdbill N95s as well as the KN95 masks are considered to be contaminated after seeing one patient, so they can’t be reused from room to room.

What do you think about triaging patients for ventilators?

In general, regardless of COVID-19, when a patient with cancer develops respiratory difficulty in the hospital and requires a ventilator, the outcome is fairly poor. In such cases, the chances of getting out of the hospital are probably less than 20% to 30% for patients with stage IV cancer who wind up on a ventilator.

Patients do not usually know these data, so our palliative care team has mobilized to discuss this information with patients who have end-stage cancer. Of course, we are obligated to follow patients’ wishes, but many patients are not aware of what it means to go on a ventilator with late- or end-stage cancer.

The other complication during COVID-19 is that Columbia, like nearly all New York City institutions, has a no-visitor policy. Going into the hospital with late- or end-stage cancer and winding up isolated from family and potentially on life support is a decision that needs to be made very carefully. Dr. Dawn Hershman and Dr. Craig Blinderman from palliative care have both done a good job presenting these issues and helping drive communication and patient outreach. 

DISCLOSURE: Dr. Drake owns stock or other ownership interests in Compugen, Harpoon Therapeutics (sponsor of one of the trials discussed), Kleo Pharmaceuticals, Tizona Therapeutics, Urogen Pharma, and Werewolf; has served as a consultant or advisor to AstraZeneca/MedImmune, Bayer, Bristol-Myers Squibb, Compugen, F-Star, Ferring, Genocea Biosciences, Janssen Oncology, Kleo Pharmaceuticals, Merck, Merck/Serono, Pfizer, Pierre Fabre, Roche/Genentech, Shattuck Labs, Tizona Therapeutics, Urogen Pharma, and Werewolf; has received institutional research funding from Bristol-Myers Squibb; holds institutional patents licensed to Bristol-Myers Squibb and Potenza Therapeutics; and has been reimbursed for travel, accommodations, or expenses by Merck Sharp & Dohme, Pfizer, and Roche/Genentech.


1. Obradovic AZ, Dallos M, Zahurak ML, et al: T-cell infiltration and adaptive Treg resistance in response to androgen deprivation with or without localized vaccination in localized prostate cancer. Clin Cancer Res. March 15, 2020 (early release online).