Razelle Kurzrock and Shumei Kato on Immunotherapy and Hyperprogression of Disease


Drs. Kato and Kurzrock discuss the potential reasons why some cancer patients experience hyperprogression of disease while on immunotherapy treatment.

Oncology (Williston Park). 32(4):150-1, 163.

Shumei Kato, MD

Razelle Kurzrock, MD

In August 2017, Drs. Kurzrock and Kato and colleagues published a study[1] analyzing potential genetic biomarkers of so-called “hyperprogressors”-cancer patients who had been treated with an anti–programmed death 1 (PD-1) or anti–programmed death ligand 1 (PD-L1) antibody and/or other types of immunotherapy. These patients experienced what the authors described as the unexpected result of faster tumor growth and worsening disease following treatment. Similar observations of hyperprogression had been documented and published in 2017 by clinical researchers at the Institut Gustave Roussy in Paris.[2]

1. Dr. Kato, could you tell us about your initial experience with this phenomenon of hyperprogression?

DR. KATO: We had a couple cases that were puzzling; we didn’t know whether the patients were having rapid progression from immunotherapy, or just regular progression or pseudoprogression. The first case was a 65-year-old woman with endometrial stromal sarcoma who had received several lines of therapy and had indolent disease with gradual progression. We gave her nivolumab because there were some efficacy data for the drug in sarcoma. About a month after initiation of nivolumab, she developed multiple new large masses in the abdomen, and existing liver metastases were growing rapidly. The second case was a 73-year-old man, a bladder cancer patient who actually had a marker for immunotherapy response-high tumor mutation burden. At that time, atezolizumab, an anti–PD-L1 antibody, had just been approved for his type of cancer, so we initiated treatment for him with the drug. However, within a few weeks of starting therapy, the patient’s clinical condition rapidly declined. Repeat scans, including a CT scan, showed, unfortunately, multiple emerging liver metastases. Both patients showed rapid tumor growth that was not in keeping with the slower pace of the tumor progression before immunotherapy. Those were our first experiences with hyperprogression. Back then we were not really sure what was going on with these patients.

2. Dr. Kurzrock, was any pattern emerging in the handful of patients whose disease seemed to be getting worse following treatment, as Dr. Kato just described?

DR. KURZROCK: Yes, it is very interesting about the pattern and how this whole phenomenon emerged for us. As Dr. Kato mentioned, there were initially two patients-Dr. Kato and I share patients in the Rare Tumor Clinic-and these two patients had fairly slow-growing disease before beginning treatment with a checkpoint inhibitor. But then, their disease exploded when they were on the checkpoint inhibitor. It was very curious because this had not been reported before, and Dr. Kato and I discussed whether this acceleration or fueling of disease was real. Then Dr. Kato came back to me and asked if I had noticed that they both had MDM2 amplification. MDM2 amplification is a specific genomic abnormality that is found in only about 5% of patients. And I thought, well, that is a little strange, what a coincidence. And then, just to tell you how this evolved, I was talking to one of my colleagues at MD Anderson Cancer Center, a physician who has also treated a lot of patients with immunotherapy, and I asked him if he had ever seen patients who looked like they had gotten worse on immunotherapy. He said that he had, and I asked, “Why didn’t you publish this?” His reply was, “I didn’t think anyone would believe us.” I asked him whether he did the genomics on these patients, and he told me he had one patient who came to mind immediately who had undergone a genomics analysis. He sent me the analysis of the patient’s tumor and the patient also had an MDM2 amplification. (I had not shared with him the fact that our two patients had MDM2 amplification.) At that point, I realized that it could not be a coincidence because MDM2 amplification occurs in 5% of patients. That led to the study in which Dr. Kato and I analyzed all of the patients who had been treated with immunotherapy at Moores Cancer Center and whose data were available to us through our master data analysis protocol, and found that both MDM2 amplification and EGFR alterations correlated with this phenomenon of hyperprogression in multivariate analysis.

3. Dr. Kurzrock, could you describe the study that you and Dr. Kato and colleagues did in a bit more detail? How many patients were included, and did you look at other genetic alterations besides the two you just mentioned?

DR. KURZROCK: There were about 150 patients total, and I want to give credit to Dr. Kato for doing the bulk of the work for that study. We started out looking at any patients who had progressed in less than 2 months, which would be considered rapid progression. We know that physicians don’t want to take patients off immunotherapy because there have been so many positive results; therefore, if patients were taken off before 2 months of therapy had been completed, we suspected that the physician’s hand had probably been forced. Then we looked at all of the genomic alterations and we performed a multivariate analysis, which showed MDM2 amplification. The other alteration that came from the multivariate analysis was an EGFR alteration. Dr. Kato then went back and pulled all of the patients’ films, so that we could examine the rate of growth of their tumors. We had very strict criteria because one of the reactions we expected was people asking, “How do you know that these patients’ tumors weren’t growing very quickly even before the immunotherapy?” Because of this, we elected not to include any patient unless we had films from 2 months before they started immunotherapy and at baseline, so we could see that the pace of growth had indeed increased at least twofold-and in some patients, the growth had increased 35- to 40-fold.

4. Dr. Kato, is there anything else you would like to add about the major findings of the work?

DR. KATO: I think one of the key things was to compare the images pre-immunotherapy, meaning not just looking at the baseline scan, but also comparing 2 months before starting the immunotherapy to baseline and then after starting the immunotherapy, to figure out the pace of progression.

5. Finally, it seems that there are many unanswered questions. Are there any that you and your colleagues are currently addressing on this phenomenon of worsening disease following immunotherapy?

DR. KURZROCK: Yes, there are a few important things to mention. The first is that we were almost afraid to submit the report because, as our colleague at MD Anderson Cancer Center said, we also felt that it might not be believed. Then the Institut Gustave Roussy published their report[2]-I want to give them credit for publishing that first report and I also want to give credit to the journal Clinical Cancer Research for publishing it-and they found that 9% of their patients were showing hyperprogression, which we think is a rate that was very similar to our patient group. What we added that they did not have is the genomic correlate. What is very interesting and encouraging is that there was an independent abstract[3] presented at the recent European Society for Medical Oncology Congress that showed that the exact same genomic correlates-MDM2 amplification and EGFR alteration-are associated with this phenomenon of hyperprogression.

I think that there are still a few very important unanswered questions. The first is that we don’t understand the mechanisms, and being scientists, we’d like to understand the mechanisms. We found something that we think is very important for patients but we don’t understand how that occurs, so we are working with our lab scientists to do some fundamental research to understand the mechanisms. The other unknown is whether this occurs in all patients who have these alterations. I think that patients with MDM2 amplification and EGFR abnormalities are at risk, but I wouldn’t say that we think that this is universal among these patients. Another unknown is whether there is a way to overcome this problem. For instance, could one give a checkpoint inhibitor together with an MDM2 inhibitor, and would that mitigate whatever effect the MDM2 amplification might have on response to the checkpoint inhibitor? So, yes, I think this raises a lot of new questions for us that we need to answer.

Financial Disclosure: Dr. Kurzrock receives research funding from Foundation Medicine, Genentech, Guardant Health, Incyte, Merck, Pfizer, Sequenom, and Serono, as well as consultant fees from Actuate Therapeutics, Loxo Oncology, and Sequenom, and speaker fees from Roche. She is a consultant for Genentech, and has an ownership interest in CureMatch, Inc. Dr. Kato has no significant financial interest in or other relationship with the manufacturer of any product or provider of any service mentioned in this article.


1. Kato S, Goodman A, Walavalkar V, et al. Hyperprogressors after immunotherapy: analysis of genomic alterations associated with accelerated growth rate. Clin Cancer Res. 2017;23:4242-50.

2. Champiat S, Dercle L, Ammari S, et al. Hyperprogressive disease is a new pattern of progression in cancer patients treated by anti-PD-1/PD-L1. Clin Cancer Res. 2017;23:1920-8.

3. Singavi AK, Menon S, Kilari D, et al. Predictive biomarkers for hyper-progression (HP) in response to immune checkpoint inhibitors (ICI): analysis of somatic alterations (SAs). Presented at the ESMO 2017 Congress. Abstract 1140PD.

Related Videos
Thomas Marron, MD, PhD
PD-1 protein bound to PD-L1