Skip to main content

12-12-2018 | Immunotherapy | Feature | Article

Hyperprogression as a response to immunotherapy: ‘A serious downside’

In light of recent reports of hyperprogression in response to treatment with cancer immunotherapy, medwireNews spoke to medical oncologist Razelle Kurzrock (University of California, San Diego, La Jolla, USA) about the phenomenon, the implications for therapy, and future directions for research.

Immunotherapy drugs, especially those targeting the programmed cell death protein 1 (PD-1) pathway, have changed the treatment landscape of various tumor types, such as lung cancer and melanoma. These agents elicit durable responses and have better tolerability than standard chemotherapy. However, a few studies – the most recent one published in JAMA Oncology [1] – have shown that some patients can have an atypical and rather adverse response to these drugs, termed as hyperprogression, which thus far appears to be a response pattern specific to treatment with agents targeting PD-1 and its ligand PD-L1.

Although the definition of hyperprogression varies across studies, it describes a situation where tumor growth is greatly accelerated and accompanied by clinical deterioration following therapy.

medwireNews spoke to Razelle Kurzrock, a medical oncologist from the University of California, San Diego in La Jolla, USA, who has been charting the hyperprogression phenomenon from the beginning.

We long suspected…

From looking at the progression-free survival (PFS) curves of immunotherapeutic interventions, Kurzrock said that they suspected for a while that such a phenomenon existed. She explained that generally when you compare two drugs that differ in their effect on PFS, they tend to be proportionately different; in other words, if one drug is better than the other, then it is consistently better from the beginning to the end.

But this is not what you see in the trials comparing immunotherapy with chemotherapy. You see “very interesting phenomena,” such that at the very beginning, the immunotherapy appears to be worse than chemotherapy, but the curves cross after the first couple of months and the immunotherapeutic agent comes out on top, explained Kurzrock.

She said that they suspected for a long time that the reason for this pattern is hyperprogression – “the first group of patients that fall off are hyperprogressors,” and once they are lost, you start to see the effects in the subset of patients that do very well on immunotherapy.

A real phenomenon

As far as Kurzrock is aware, the JAMA Oncology study by Benjamin Besse and colleagues is the fourth to report on hyperprogression as a response to PD-1 or PD-L1 inhibitor therapy, following research by her own team and two other research groups [3, 4]. All have found patients – “a small subset but not negligible” – where immunotherapy appears to accelerate the pace of their progression, she noted, with rates of 9–29% reported thus far. She added that although the underlying mechanisms remain unknown, it appears to be “a real phenomenon”

And although the Besse et al study focused on patients with non-small-cell lung cancer, other research has documented hyperprogression in other cancers, such as melanoma and head and neck cancer, and therefore this phenomenon is “absolutely not restricted” to a single tumor type, said Kurzrock.

Identifying the hyperprogressors

Given that hyperprogression is such an atypical response, identifying the patients who are likely to experience it becomes crucial.

Kurzrock told us that her team has identified two potential genomic markers so far – MDM2 amplification, which was very strongly associated with hyperprogression, and EGFR aberrations, which showed a weaker association.

She noted that the Besse et al study did not validate the association between hyperprogression and EGFR mutations, as none of the patients with EGFR-mutated disease in their cohort had hyperprogressive disease, but she believes that the association exists and needs to be assessed further. There may also be other factors that at play as well; “I don’t want to pretend that we know everything,” she commented.

In the meantime, Kurzrock suggests that one option could be to run genomics on patients prior to initiating anti-PD-1 or anti-PD-L1 therapy, and if they test positive for an MDM2 amplification or EGFR alteration, then “probably be pretty cautious about treating the patient or not treat the patient at all.”

Hyperprogressor or pseudoprogressor?

Another type of atypical response that has been observed with immunotherapy is pseudoprogression – this is when the tumor appears to grow on the initial scan post-immunotherapy but ultimately responds to treatment.

Since hyperprogressors and pseudoprogressors will both initially show tumor growth after receiving immunotherapy, we asked Kurzrock if there is a way to tell these patients apart as you would not want to stop treating a patient who is showing pseudoprogression.

“You’ve hit the nail on the head,” she said, noting that they have some experience of such scenarios in the clinic. She described a patient – “a classic problem in this regard” – who appeared to be a good candidate for immunotherapy on the basis of having a high tumor mutation burden. But at 6 or 7 weeks after starting treatment, their scans looked much worse – the tumor appeared to have grown.

Now we know that this can happen in the case of pseudoprogression, where the tumor grows due to the high degree of infiltration by immune cells, said Kurzrock. “It looks worse on the scans but then the patient gets better because the immune cells just eliminate the tumour and then the next scan is better.”

She pointed out, however, that there are no good tools at the moment to distinguish between hyper- and pseudoprogression, but one thing that physicians can do – and indeed is the strategy they employed with this patient – is to ask the patient how they feel.

Patients experiencing pseudoprogression will often tell you that they feel much better, while those with hyperprogressive disease will say they feel worse and indeed, they will look worse, explained Kurzrock. She stressed that this is not an exact science, but until we have better markers, “just asking the patients might be something that is helpful.”

Looking to the future

In terms of future work, Kurzrock highlighted three key avenues of research. First, she believes it is important to validate the genomic markers that her team has discovered and also to identify any other markers that may exist.

The second aspect is to understand the biology of hyperprogression, she noted, adding that they are already working with basic scientists to unravel the underlying mechanisms.

And finally, Kurzrock thinks that it is critical to find markers that can distinguish hyperprogression from pseudoprogression, and her team has already made a start. She said that their latest research has revealed a potential role for circulating tumor (ct)DNA in this regard. The levels of ctDNA track with hyper- and pseudoprogression, she said, such that ctDNA levels increase in the former scenario and decrease in the latter.

She noted, however, that although the evidence suggests that ctDNA may be a helpful marker, their study had a small sample size and ctDNA needs to be examined in larger cohorts.

‘The best drugs ever invented for cancer’

In conclusion, Kurzrock emphasized that in spite of the possibility for hyperprogression, “I still think these immunotherapy drugs are the best drugs ever invented for cancer.”

“They have a downside, and it's a serious downside, but I do want to put it in perspective that the upside is much bigger,” she commented, highlighting that the proportion of patients that respond is higher and also that some of the responses to these drugs “are really remarkable.”

“We've seen patients who were on their way to hospice who are now in complete remission for 2 to 3 years – we never saw that with any other type of therapy,” Kurzrock said.

Nevertheless, she cautions that immunotherapeutic drugs should not be given indiscriminately. We have markers that identify patients who are likely to respond, and hopefully we will soon also have markers to identify those most likely to hyperprogress, she concluded.

By Shreeya Nanda

medwireNews is an independent medical news service provided by Springer Healthcare. © 2018 Springer Healthcare part of the Springer Nature group

Image Credits