Immunotherapy Linked to Hyperprogression of NSCLC and Other Cancers

December 21, 2017
Jennifer Fink

Targeted Therapies in Oncology, November 2017, Volume 6, Issue 11

Hyperprogressive disease (HPD) after immunotherapy treatment may not be as rare of a phenomenon as previously thought. A recent multicenter, retrospective analysis of 242 patients with advanced non&ndash;small cell lung cancer&nbsp;(NSCLC) found that 16% of patients developed hyperprogression during anti&ndash;PD-1/ PD-L1 treatment.<sup>1</sup> The study, which was presented at the 2017 ESMO Annual Congress, is one of the latest to highlight the risk of hyperprogression.

1The study, which was presented at the 2017 ESMO Annual Congress, is one of the latest to highlight the risk of hyperprogression.

Awareness of hyperprogression—a rapid increase in tumor growth after initiation of immunotherapy—has been increasing since a 2016 French study first defined hyperprogression in a population of patients receiving anti—PD-1/PD-L1 therapy for melanoma, lung, renal, colorectal, head and neck, breast, brain, cervical, endometrial, and other cancers.2That study, which was published inClinical Cancer Research, identified hyperprogression in 12 of 131 patients, for an incidence of 9%. Additional research studies suggest that rapid tumor growth during immunotherapy is likely more common than that, particularly in certain patient populations.

Understanding the definition, incidence, and pathogenesis of hyperprogression will ultimately help clinicians provide better care to patients with cancer. “We know that immunotherapy doesn’t work for everyone. We need to do better,” said Charu Aggarwal, MD, MPH, an assistant professor of hematology/ oncology in the Perelman School of Medicine at the University of Pennsylvania, who was instrumental in opening phase I clinical trials of pembrolizumab (Keytruda) in lung cancer at the University of Pennsylvania. A key part of “doing better,” she said, will be learning how to accurately identify which patients are likely to benefit from immunotherapy, and which are not.


The French study described HPD as a Response Evaluation Criteria in Solid Tumors (RECIST) progression at the first evaluation and a 2-fold or greater increase in tumor growth rate from baseline (before treatment with immunotherapy).2 Similarly, an Annals of Oncology article from April 2017 about hyperprogression during anti—PD-1/PD-L1 therapy in patients with head and neck cancer defined hyperprogression as a tumor growth kinetics ratio equal to or greater than2(FIGURE).3And an August 2017 study by Shumei Kato, MD; Razelle Kurzrock, MD; and colleagues, used a 3-point definition of hyperprogression4:

1. Time-to-treatment failure (TTF) of less than 2 months

2. Greater than 50% increase in tumor burden compared with preimmunotherapy imaging

3. Greater than or equal to a 2-fold increase in progression pace

“It’s important to not only compare imaging from right before immunotherapy and after immunotherapy, but to also compare with scans taken about 2 to 3 months before, to look at the pace of progression,” said Kato, an assistant clinical professor of medicine at the University of California, San Diego (UCSD) School of Medicine. Kurzrock is the chief of the Division of Hematology and Oncology, senior deputy director of clinical science, and director of the Center for Personalized Cancer Therapy and Clinical Trials Office at UCSD. Patients who exhibit hyperprogression may have relatively slow or stable tumor growth prior to immunotherapy treatment, followed by a rapid increase in tumor size after initiation of immunotherapy.

Although researchers and clinicians haven’t yet agreed upon a universally accepted definition of hyperprogression, evidence of its existence has been established via a handful of studies from around the world. “A name has been given to an entity that was previously undescribed,” Aggarwal said in an interview withTargeted Therapies in Oncology&trade;.


While the first published study of hyperprogression reported an incidence of 9% in a population of patients receiving anti—PD-1 or anti–PD-L1 monotherapy for a variety of cancers,2researchers speculated that the actual rate of hyperprogression could be higher, as 18 patients were excluded from evaluation because they exhibited clinical progression before tumor evaluation. This study found a significant difference between HPD status and age. Patients who experienced hyperprogression were older than patients who did not (66 vs 55 years;P= .007), and 19% of patients (7 of 36) older than 65 presented with HPD compared with 5% of patients (5 of 95) younger than 64 years (P= .018).

3 That incidence rate was based on an analysis of the medical records of 34 patients, of which 10 exhibited hyperprogression. The median age was 63 years, and researchers noted that “previous irradiation might play a role, since almost all cases of hyperprogression occurred in patients who had at least a locoregional recurrence in an irradiated field. Mechanisms behind this hypothesis are unknown.”

The multicenter study of 242 patients with advanced NSCLC found a 16% rate of hyperprogression during anti−PD-1/PD-L1 treatment.1Compared with the non-HPD population, patients with HPD had significantly lower median progression-free survival (1.4 vs 4.9 months;P<.001) and median overall survival (3.4 vs 17 months;P<.001).

Recent research suggests that the prevalence of hyperprogression may be related to the underlying genomic make up of tumor cells. A single-center retrospective analysis of hyperprogression after immunotherapy treatments of a variety of cancers found hyperprogression rates as high as 67% in patients withMDM2amplification, and in 20% of patients withEGFRalterations.4


The link betweenMDM2amplification,EGFRalterations, and hyperprogression on immunotherapy treatment was revealed after Kato and Kurzrock noticed that several of their patients experienced rapid disease progression after initiation of immunotherapy. One was a 65-year-old woman with endometrial stromal sarcoma; she had undergone “several lines of treatment but was having slow-growing disease,” Kato said, including progression of liver metastases over 6 months with targeted therapy. She wanted to try immunotherapy and was switched to nivolumab (Opdivo) and stereotactic body radiation therapy. Within 2 weeks, she was complaining of increased abdominal pain and swelling, Kato said.

A CT scan revealed growth of the existing liver masses, and a grapefruit-sized mass in the abdomen. “We were scratching our heads, wondering if this was some kind of pseudoprogression,” Kato said. “We did a biopsy, and it was full of cancer.”

After a second patient exhibited an increase in clinical symptoms and liver metastases after initiation of atezolizumab (Tecentriq), despite the fact that his tumor had a high mutational burden and therefore was expected to respond well to immunotherapy, Kato and Kurzrock decided to look at genomic data.

“What we found is that, among the multiple genome alterations those 2 patients had, they had one in common:MDM2amplification,” Kato said. Intrigued, they analyzed the medical records of all patients with stage IV cancers who received CTLA-4, PD-1/PD-L1 inhibitors and other investigational immunotherapy agents, and had comprehensive genomic analyses

(n = 155) completed at their center between March 2011 and July 2016.

“Because we do genomics routinely here at UCSD, we looked at all patients who received some type of immunotherapy and also had a genomic analysis,” Kato said. They discovered that alterations in several genes—namely,TERT,PTEN,NF1, andNOTCH1— were associated with favorable clinical outcomes, while others (notably,EGFR,MDM2/4, andDNMT3A) were associated with a TTF of less than 2 months (TABLE).4After multivariate analysis, onlyEGFRandDNMT3Aalterations andMDM2amplifications were significantly correlated with a TTF of less than 2 months.4

Evidence of hyperprogression was apparent in 4 of 6 patients (67%) withMDM2family amplifications. Increases in tumor lesion size, compared with preimmunotherapy treatment, ranged from 55% to 258%. Hyperprogression was also noted in 2 of 10 patients (20%) withEGFRalterations, with lesions increasing by between 53.6% and 125%. Of note, all the patients who experienced hyperprogression were treated with anti—PD-1/ PD-L1 monotherapy.4

The apparent connection betweenEGFRalterations and hyperprogression during immunotherapy treatment is interesting, given that previous research and clinical experience have already demonstrated that lung tumors featuringEGFRmutations are more likely to shrink in response to targeted therapy than immunotherapy.EGFRactivation is also known to be associated with upregulation of PD-1, PD-L1, and CTLA-4.

BothMDM2andMDM4inhibit the p53 tumor suppressor; yet, to date, no one knows exactly howMDM2family alternations might trigger hyperprogression. The researchers noted that immune checkpoint inhibitors can lead to elevated levels of interferon-gamma, which activates JAK-STAT signaling. This results in an increase in interferon regulatory factor-8 expression, which inducesMDM2expression.

“It is conceivable that this cascade may not have significant impact whenMDM2is not amplified; however, in the presence ofMDM2amplification, hyperexpression could occur. Other hypotheses are also plausible, including the involvement of a gene that sits on theMDM2amplicon and is coamplified with it,” the authors wrote in the study.

Aggarwal called the link between MDM2 and hyperprogression “an interesting observation and a very novel discovery.”


Though much remains to be learned about hyperprogression, the available research suggests that caution is warranted when using immunotherapy to treat patients withMDM2amplification orEGFRmutations. “I don’t think we can strongly say that we should avoid putting patients on immune checkpoint inhibitors if they have anMDM2amplification,” Kato said. “At the University of California, San Diego, we are definitely careful when we use immune checkpoint inhibitors for these patients. We may recommend combination treatment. But we don’t have the data that show us how to prevent a patient from having a quick progression.”

Of course, it’s not yet standard protocol to perform genomic sequencing on patients before initiating immunotherapy. “We do it standardly for all of our patients with nonsquamous non—small cell lung cancer, but for patients with squamous cell lung cancer, it’s not standard to do it,” Aggarwal noted. That may change, though, as researchers learn more about genetic alterations and responses to immunotherapy.

It’s also not current practice to measure the pace of tumor

progression prior to initiation of immunotherapy, or to measure progression or regression via serial CT scans. Although serial CT scans provide evidence of hyperprogression in research studies, there’s no need to increase radiographic monitoring of patients receiving immunotherapy. “

“If you look at the literature of hyperprogressors, they all clinically manifested before they were radiographically picked up,” Aggarwal said. “Clinical symptoms should precipitate the need for scans. I don’t think the rates of hyperprogression justify the use of radiographic analysis for all patients at specified time points.”

If hyperprogression is evident, immunotherapy treatment should be stopped. However, management of hyperprogression will likely evolve as more is learned. For instance, MDM2 inhibitors are currently in clinical development; in the future, combination treatment with an MDM2 inhibitor and immunotherapy could possibly limit hyperprogression.


To date, it is unclear whether the various reported rates of hyperprogression are the result of small study sizes, differing responses to immunotherapy based on cancer histology or immunotherapy medication, or something else. Researchers and clinicians are continuing to work toward an understanding of why some patients experience full regression with immunotherapy therapy treatment, while other see rapid advancement of their underlying disease.

Kato and Kurzrock are currently working with laboratory collaborators to better understand the mechanisms by whichMDM2amplification may cause hyperprogression. That kind of work will help to elucidate the link betweenMDM2amplification and hyperprogression, and may point the way toward more effective treatments.

“To move forward, we need to understand how exactly it leads to this rapid progression,” Kato said. “By understanding this, we may be able to overcome hyperprogression.”

Aggarwal and colleagues are currently working on a prospective clinical trial to correlate prospective gene sequencing monitoring with either response or progression to single-agent pembrolizumab, she said. Such work will add to the body of knowledge regarding immunotherapy, and may point to important biomarkers clinicians can use to guide treatment.

Additional research is urgently needed, though. “We have a lot to do in terms of understanding and picking out which patients are predisposed to hyperprogression,” Aggarwal said.


  1. Ferrara R, Caramella C, Texier M, et al. Hyperprogressive disease (HPD) is frequent in non-small cell lung cancer (NSCLC) patients (pts) treated with anti PD1/PD-L1 monoclonal antibodies (IO). Presented at: 2017 ESMO Congress; September 8-12, 2017; Madrid, Spain. Abstract 1306PD.
  2. Campiat S, Dercle L, Ammari S, et al. Hyperprogressive disease (HPD) is a new pattern of progression in cancer patients treated by anti-PD-1/PD-L1.Clin Cancer Res.2016;23(8):1920-1928. doi: 10.1158/1078-0432.ccr-16-1741.
  3. Sa&acirc;da-Bouzid E, Defaucheux C, Karabajakian A, et al. Hyperprogression during anti-PD-1/PDL1 therapy in patients with recurrent and/or metastatic head and neck squamous cell carcinoma.Ann Oncol.2017;28(7):1605-1611. doi: 10.1093/annonc/mdx178.
  4. 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(15):42424250. doi: 10.1158/1078-0432.ccr-16-3133.