Outcomes Related to Cancer Treatments in Patients with COVID-19 from the CCC19 Registry Analysis- Ali Khaki, Jeremy Warner & Petros Grivas

October 6, 2020

The COVID-19 and Cancer Consortium (CCC19) is a consortium of over 120 cancer centers and organizations coming together to collect data about patients with cancer who have been diagnosed with COVID-19. Ali Khaki, MD, Jeremy Warner, MD, MS, FAMIA, FASCO, and Petros Grivas, MD, Ph.D., join Alicia Morgans, MD, to discuss how the CCC19 consortium has evolved since first being established earlier this year and highlight two CCC19 registry analysis presented at the 2020 virtual European Society of Medical Oncology (ESMO) annual meeting. They discuss outcomes related to systemic cancer treatments within one year of lab-confirmed SARS-CoV-2 infection in an expanded cohort that was examined.  ANd they share findings on the factors associated with high mortality in the largest cohort of patients with cancer and COVID-19 that were systematically and comprehensively identified and assessed. 


Ali Khaki, MD, Hematology/Oncology Fellow, Fred Hutchison Cancer Research Center, University of Washington, Seattle, Washington, United States

Jeremy Warner, MD, MS, FAMIA, FASCO, Associate Professor of Medicine, Division of Hematology/Oncology, Associate Professor of Biomedical Informatics, Vanderbilt University

Petros Grivas, MD, Ph.D., Associate Professor, Clinical Director of Genitourinary Cancers Program, University of Washington, Associate Member, Clinical Research Division, Fred Hutchinson Cancer Research Center.

Alicia Morgans, MD, MPH Associate Professor of Medicine in the Division of Hematology/Oncology at the Northwestern University Feinberg School of Medicine in Chicago, Illinois.

Read the Full Video Transcript

Alicia Morgans:  Hi, my name is Alicia Morgans and I'm a GU medical oncologist and associate professor of medicine at Northwestern University. I'm so excited to have here with me today, friends, and colleagues, Dr. Petros Grivas, who's an associate professor of medicine and a GU medical oncologist at the University of Washington. Dr. Jeremy Warner, who's an associate professor in the department of biomedical informatics and a hematologist at Vanderbilt University, and Dr. Ali Khaki, who is a senior and super fellow at the University of Washington. Thank you so much for being here with me today.

Petros Grivas:  Thank you so much, Alicia, for having us [crosstalk].

Jeremy Warner, Ali Khaki:
  Good to be here.

Alicia Morgans:  Wonderful. So I wanted to talk with all of you about a really important and exciting presentation that you've had at ESMO, the virtual meeting, 2020.  We learned a lot about the CCC19 grassroots, sort of bioinformatics database on patients who have cancer and have been afflicted with COVID-19. And I'd love to hear a little bit more about how this came about, and why it's so important. Can you tell us, Dr. Warner?

Jeremy Warner:  Sure thing. So we've been now running this effort for more than six months. This started all the way back in March before COVID really arrived on the shores of the United States. I think there were fewer than a thousand cases at that point, as opposed to 7 million at this point.  But we had seen some literature coming out of China that was really limited in the conclusions it was making about patients with cancer. So the question we asked is, can we get together enough institutions to pool our resources, to pool our knowledge, and actually learn about what is happening to the patients who are getting COVID-19. So what started as really just five institutions became 14, became 124, as of the last count.

So really, really rapid growth, both in the number of institutions participating as well as unfortunately, the number of cases that we've had reported. I think recently we've gone well past 5,000. So we had reported on about the initial 1000 patients in a paper that was published in The Lancet in May, along with the presentation at ASCO. And since that time, as I said, the number of patients has gone up dramatically and we wanted to update our analysis and also add some new questions that we could answer with larger numbers. So that was the basis of our ESMO presentation.

Alicia Morgans:  Really important, especially as we are all still grappling with and trying to adapt to the new normal that we now have, living with COVID-19, Dr. Khaki, can you tell us a little bit about the presentation at ESMO this year, and what you studied and what you found?

Ali Khaki:  Yeah, so there are actually two presentations by the CCC19 that were made. One was by Dr. Wise-Draper, who talked about the association between the timing of treatment and specific treatments and outcomes for patients with cancer who had a COVID infection. And the second was by Dr. Grivas, who was looking at other prognostic factors, other clinical or demographic factors of patients, as well as an initial exploration of laboratory values and their association with 30-day mortality, with sort of SARS-CoV-2 infection. And what we found is, many of the initial factors that we had identified and presented on at ASCO, and in the initial Lancet paper. So, things like older age, male sex, progressing cancer, worse ECOG performance status, those things were all still negative prognostic factors. In addition to that, we had our first foray into laboratory markers, to see how those were associated with worse outcomes. And what we saw is patients who had a lower absolute lymphocyte count, so the lower limit is normal.

Those with a high or low absolute neutrophil count, those with a low platelet count, those with an abnormal creatinine, an abnormal D-dimer, an abnormal troponin, or an abnormal CRP, all had sort of an association with worse outcomes, worse, 30-day mortality. The way that we did this modeling, was, each one of these was put into a model independently. So we have not sort of adjusted labs for each other, but this was our first exploration. And as we sort of have more patients in our database, as well as more cases, we intend to do further investigations to delve into this further.

Alicia Morgans:  It's phenomenal because you guys have really started, as I mentioned, this grassroots database where people from around the country and actually in various places around the world are entering their data. How easy was it to sort of put all this data together, Dr. Grivas, as you were trying to do this analysis?  I can imagine that it would be challenging to get everything together, to actually do this analytic plan, and then present what you could find at this meeting.

Petros Grivas:  I absolutely agree, Alicia.  This is a gigantic effort, and I have to give kudos to Jeremy, and not because he's present, but because he has been the orchestrator of this effort and his Vanderbilt team has been the core of that. However, as Jeremy mentioned, this team effort with 124 institutions and numerous investigators across the board, that will have three countries represented, the United States with all the different geographical diversity from all the parts of the United States. We have also institutions from Canada and Spain. So, technically it's an institutional effort from multiple countries, and literally, two continents here, and three countries, as I mentioned.  This represents multiple logistical hurdles and challenges when you try to get all these people together and their regulatory hurdles. So, in the beginning, I remember we're discussing how we can streamline the process and how can we establish better user agreements, how can we make this regulatory compliant?

And then how can at each site, the IRB can approve these efforts relatively quickly, and still stick with compliance, but also avoid roadblocks, remove barriers to make this a quick effort, and generate the data as quickly as possible because the pandemic was running quickly, and not running with it. So, the effort I would say is pretty remarkable and multiple barriers can carry these multi-site collaborations, but CCC19, I think represents one example of how much and how quickly things can be done when there is a will, and there's a way, as we say, and we are able to discuss with individual IRB's, try to expedite, that if you process at individual sites, establish the user agreements. Jeremy had a number of lawyers at Vanderbilt University to try to work out these other user agreements. And this was definitely one of the most difficult tasks, but at the same time, it was a priority, and still is a priority for many of us who shifted focus to the part of our research towards COVID-19 and cancer.

And I think the results definitely reward this continued effort, but nothing can be done unless we have a big team. And the team is the answer here. And the team is from 124 institutions, three countries, and also a huge number of patients.  As Ali mentioned, we have about 5,200 cases in the last analysis that we looked at. And this is definitely something that is unprecedented. One of the highest, probably the largest consortium worldwide. We have a steering committee of, used to be 11, now we have nine members, and we work together strongly.  But again, the heart and the mind go out to all the investigators.

Alicia Morgans:  Wonderful. And I have to also congratulate you, Jeremy, just as Dr. Grivas has, and really kind of being the linchpin that has pulled everyone together. And I think there is no better person to kind of be that head that pulls everyone together, that head person because you do always need a leader, of course, with all the team around you. So congratulations on that. And I just want to ask, does any of this data actually affect your practice? I know you're a hematologist. You may not necessarily see any solid malignancies, and we'll ask Dr. Grivas and Dr. Khaki the same thing, but as a hematologist, we have heard that some of our heme malignancy patients may be afflicted even more heavily than some of our solid tumor patients.  Have any of your findings actually impacted your practice or the way that you manage your patients?

Jeremy Warner:  Yeah. So first I'll say if all this falls apart, I think we can all join the circus because we've been walking a tight rope for months. No, I mean really, I mean the idea here was to get people to donate data, but not to the point where they would just walk away because it took two hours to fill out a form, so I think in order to build this thing, we had to accept that the data coming in would be not pristine, to put it plainly, but we created a structure that it's not a one and done. So the idea is that a site can report the information they know, at that time, and then go back, fill in what they don't know, including the follow-up of course, that we don't know at the time that a case is reported.

Honestly, when we started this thing, we couldn't think beyond 90 days, to us, that was like a very long horizon, right? That anybody would have COVID-19 and then 90 days later, but now we're realizing that's obviously just the beginning of what's going to be a long-term sort of effort to understand what happens for the survivors.  But in terms of hematologic malignancy, I mean, one of my drivers to get involved, and get going with this thing was that I was concerned, especially for patients that have a lot of healthcare contact. So these days I basically only provide inpatient services. And certainly, those patients have very, very heavy contact with the healthcare system. And we still give a lot of, a lot of our chemotherapies infusional, and there are alternatives. So classic would be the dose adjusted R-EPOCH versus an R-CHOP, so you've got a five-day hospital admission versus a couple of hours in the clinic.

And so those kinds of questions were kind of forefront in my mind. Now I do say that, unfortunately, we had a bit of a false sense of security with our first publication in the Lancet because we just didn't have the power. So even though we had a thousand patients, only 20% of those were hematologic malignancies or had a hematologic malignancy, and it didn't pan out. There was no statistically significant difference.  That is no longer the case. So with our ESMO presentation, we did find that hematologic malignancy is independently a risk factor for increased mortality. I can't remember the exact hazard ratio, but I think it's in the realm of like 1.7 or so. So it's not quite double, but definitely higher than that what's for patients with solid tumors.

Now that's one thing. Now, why? I think we raise more questions than answers with our analysis so far, but one thing that greatly concerns me, as from the other presentation Dr. Khaki mentioned by Dr. Wise-Draper at ESMO, and that looked a little bit into the specific treatments that were being received and their timing. And one of the most concerning findings is, that patients receiving lymphodepleting treatment.  So the classic being Anti-CD20, like rituximab, patients who had received that treatment long enough before COVID to have basically lymphodepletion, had an extremely high rate of mortality, approaching 50%.

Now the numbers are small because you're looking at a subset of a subset of a subset, but still, it was somewhere in the order of 50 or so patients and half of them died. And technically it's statistically significant unadjusted, but it's very concerning. So I think about, I think a lot about patients who are getting treated today and what's going to happen to them when they're in the middle of flu season. And if they happen to get flu and COVID at the same time, what's going to happen. I do think about that a lot.

Alicia Morgans:  I think, and again, I, as a solid tumor oncologist I don't use rituxan at all, but when I used to use that medication, of course, we think about that being ultimately potentially lymphodepleting, but in general, it's considered relatively tolerable and something that people can get on an ongoing basis as maintenance or whatever in whatever way that they need it. So very important to find that this may be a particular vulnerability in the era of COVID. And I hope that further data with more numbers may show that that's not true, but of course, it very well could be true. And something for us to be very, very aware of with this early finding, from your data, as we continue to monitor and care for these patients over the longterm.

Jeremy Warner:  Now, just quickly, you mentioned that rituximab is, I believe, the top-selling drug period. Because it's not just used for hematologic malignancy, but a lot of autoimmune diseases.

Alicia Morgans:  Yes, exactly. I mean, it's used in benign issues as well, autoimmune issues and not without purpose, but it is used relatively easily because it doesn't generally have a high toxicity profile. So really important findings. And I do encourage anyone who's engaged in CCC19 to continue to enter their data and try to let us have some more information on these patients because over time it will be very important for us to understand, again, as Jeremy says, this is one of the highest-selling drugs, it's going to be very important for us to understand the impact of this medication on patients, whether they have cancer or not, whether they have autoimmune issues or cancers. So, Dr. Khaki, as you're thinking about your GU patients, are there any findings from this ESMO presentation or either of these ESMO presentations that really influenced your decision-making as a GU medical oncologist?

Ali Khaki:  Yeah. Good question. I don't know if there's anything specific from these findings that informed me as a GU medical oncologist, but one thing that I've learned a lot, and actually has been on top of mind for me has been sort of some of the disparities related to this. And one of the findings we did find also in the ESMO presentation was, the black race was associated with worse outcomes, which is something that's been seen in other sorts of data sets as well, both in the cancer and non-cancer population. But early in my fellowship, I spent a lot of time working at the Harbor View Medical Center here in the Seattle area, which is our sort of underserved hospital. And as I was doing data collection, I don't practice there anymore, but as I was doing data collection for CCC19, many of my old patients from Harbor View, all came up as patients I was recording into our dataset.

And so it was one of those moments where it became really real to me, how I had not seen many or any patients actually at SCCA or The University of Washington that have had COVID.  There have been some, but not very many, but many of my patients who are at Harbor View had all been infected. And so, like that sort of disparity, of just you being on the wrong side of town. And so what that means in terms of your risk of getting the infection. Fortunately, most of those patients did have good outcomes, and none of my patients had died.

But just that risk factor of getting the infection was something that I've been thinking a lot more about and trying to figure out ways that we can take care of our whole community here in the Seattle area. We've been lucky actually in Seattle where we were in the news very early on, back in March because the infection was going through some nursing homes, but it's otherwise not spread too significantly here, I'd say. And especially in North Seattle, it's not been that big an issue.  But I think definitely in South Seattle, where there are more underserved populations, it has been a bigger issue and so we're trying to figure out ways that we can make sure we can protect those populations as well.

Alicia Morgans: I think that's a really important point to make. And Dr. Grivas, as you're thinking about all of these data, both presentations, do you have any take-home messages for the folks in the audience to think about as they review your data and think about these presentations?

Petros Grivas:  Thank you, Alicia, for this important question.  It's very difficult, obviously, with a retrospective study, despite the huge magnitude of our effort to have a strong conclusion because we have made some significant associations, and as Jeremy and Ali mentioned, we have validated our priority clinical-pathological factors that were associated with higher [inaudible] mortality, in our cohort, however, association doesn't necessarily mean causation. So, my challenge is when I have a patient in front of me who needs treatment right away for metastatic urothelial cancer or significant prostate cancer what do I do? And so far, I don't think we have any clear take-home messages because we haven't done this prospectively to take patients in a randomized fashion, and say treatment now vs treatment late. My bias, personally, and then this is not supported by any high evidence data, is that patients who need treatment for cancer should not have a delay in their treatment initiation because cancer itself can cause mortality.

And I tend to treat those patients. Now, depending on where you are, in the pandemic, the level of the peak, and also depending on the control of the cancer, I can have patients who have a great disease control they are on immunotherapy for one year or so, and they have a great response.  This is under control, these patients can potentially get a treatment break and keep them away from the cancer center, especially during the peak of the pandemic. So, I think it's in the medicine individualized discussion with the patient one-to-one, based on the medical comorbidities, performance status, acuity of need to treat, alternative treatment options, treatment cancer type, all those factors. And I think if anything, our work with Jeremy and all consortium that we have at CCC19, is adding to this discussion, is contributing data sets to inform in a substantially scientific way.  This discussion.

But I don't think we can say black or white, treat no treat. I think this is a very, very elaborate discussion with the patient. We have shown for sure that patients who are having progressive cancer, who have poor performance status, who have multiple medical comorbidities, older patients, male patients, those, as Ali mentioned, with black race, history of smoking, all of these are negative prognostic factors as well, hematological malignancies, more than one cancer. And also we showed that any recent therapy given within three months was associated with higher mortality. Now, the question is, if these patients already are sicker, or frailer, could potentially die anyway, and we do not have any causative relationship proven here, it's an insignificant association, but I think that the treatment decisions have to be taken on one to one basis in the clinic.

Alicia Morgans:  Well, definitely more work to be done and individualized decisions need to be made for sure, but you guys are continuing to acquire data, continuing to of course do analysis. And I'm sure we will hear more updates as time goes on. Dr. Warner, are there specific take-home points that you would put out there for the listeners as they think about the message of CCC19?

Jeremy Warner:  I think patients with cancer who are on treatment have got to avoid getting infected. I mean, this is very clear how, obviously, that's easier to say than to actually always make happen. And I think a lot of infection comes through the family, not necessarily through healthcare settings. So I think practicing those fairly simple, social distancing, masking, thinking about vaccines when they're available, definitely getting the flu vaccine and others, and just minimizing contact with the healthcare system. I think these are of paramount importance. I do think that I mean, I agree with Dr. Grivas, I mean patients who need treatment, need treatment. I think the question is if you've got equipoise between, say a chemotherapy regimen and immunotherapy regimen, which one do you go for first? If you're thinking of combining, if you're thinking of giving a dose-dense regimen, is that wise at this point? As we just mentioned, neutropenia is a factor associated with increased mortality.

I don't believe the virus tends to cause neutropenia, so we think that there's possibly an interaction there with cytotoxic chemotherapy. So, another thing to consider. I think as a clinician obviously would want to follow the guidelines that are sort of influx. So, that's hard, but ASCO's working very hard. Others are working very hard. We at CCC19 are actively collaborating with ESMO through the ESMO co-care consortium, which is collecting information from the rest of the world. So I think there's power in numbers as well. And especially looking into some of these granular questions, like not just about a type of therapy, but a specific drug, questions like that we can start to get to once we have sufficient numbers. So I guess the last thing I'll say to your listeners who might not actually be part of the consortium is, consider joining.

We have tried to make it as easy as possible. There are some legal hoops, like we mentioned in the beginning, the data transfer agreement, and so forth, but they're not too onerous. And we'd especially like to hear from more community practices. We do have about, I think, 800 or about a thousand patients at last count that have been reported in from community practices, but that compares to about 4,000 from the academic medical center type of institution. So, I do think we're getting a bit of a biased sample, it's inevitable when you're undertaking an effort like this, but certainly, we'd like to think about getting more reports from community practices.

Alicia Morgans:  Absolutely. So, any contribution that can be made from community practices starting now is still okay, and in fact, it's great. So please go onto the CCC19 website. We will have a link with this podcast so that you can click on it, get engaged if you want to, and certainly stay abreast of all of the findings, which I think are really useful in clinical practice. And then final words from our super senior fellow, Dr. Khaki, what would your message be to folks who are thinking about SARS-CoV-2, COVID-19 infections with cancer? What is your message to folks?

Ali Khaki:  Yeah, I mean, I think the first thing I'd say is with CCC19, I've just been inspired by the entire economic research community, especially working with Dr. Grivas and Dr. Warner. they've been very gracious leaders of this project, along with the other members of the steering committee. They've been very welcoming to junior faculty or trainees and sort of made it a very hospitable environment for us to be involved in this project. The whole foundation of this started with a medical resident at the University of Connecticut, who's now a fellow at Mayo Clinic, Aakash Desai So, this comes from very humble roots. And I think that through the first six months the leadership has maintained that sort of tradition of trying to honor even junior members of the team, which has been inspiring and fun to be part of.

And then, I think in terms of sort of just reiterating what Jeremy said, I think that for those out there who are still considering, as clinicians, who have patients, please share them with us, we'd love to continue to grow this. I think that we have just begun to understand some aspects of what are the risk factors for patients and the more patients we have, the more we can learn as a community, and that will be better for all of our patients. And so I think that between the first thousand and now 5,000 we've made some progress and to hopefully the next 5,000 we'll learn even more. So, come join us.

Alicia Morgans:  Wonderful closing thoughts. Thank you so much, Dr. Grivas, Dr. Khaki, and Dr. Warner. I sincerely look forward to learning more from CCC19. Thank you so much for sharing all that you know from your wonderful dataset, thinking about COVID-19, that pandemic, and cancer patients, we really appreciate you, and we appreciate your time.

Jeremy Warner:  Thank you.

Ali Khaki:  Thank you.