After more than 2 decades of change in the epidemiology of oropharyngeal cancers, the choice of optimal treatment for patients with human papillomavirus–positive disease remains elusive.
Nabil F. Saba, MD
After more than 2 decades of change in the epidemiology of oropharyngeal cancers, the choice of optimal treatment for patients with human papillomavirus (HPV)—positive disease remains elusive. Investigators are testing strategies that could reduce the intensity of treatment for patients who are HPV positive and lessen the often serious long-term adverse events (AEs) while maintaining efficacy.
Clinical trial results available thus far indicate that a subgroup of patients who are HPV positive and have good prognoses may be safely treated with a lower-than-standard radiation dose after chemotherapy. Other studies that are testing various combinations of surgery, chemotherapy, radiation, and immunotherapy, as well as strategies for identifying patients who are eligible for less intense treatment, are expected to report out over the next few years.
Although some clinicians have individually started to deintensify treatment of their patients who have HPV-positive oropharyngeal squamous cell carcinoma (OPSCC), experts say the research has not yet established solid guidelines for practices generally. The question of how best to approach deintensification remains controversial, and investigators say more high-quality trials are necessary to make clear which new therapy sequences add benefit and in which patients.
“This is a work in progress, and I don’t think we’ll have an agreed-upon gold-standard deintensification method for another 10 years, maybe. But we’re heading in that direction,” said Nabil F. Saba, MD, director of the Head and Neck Medical Oncology Program of Winship Cancer Institute at Emory University in Atlanta, Georgia. I use the analogy of Hodgkin disease, where we started treating this with very intensive radiation decades ago, the deintensification took place, and then people moved away from high-dose radiation to chemotherapy. This process took a significant time and effort and it continues to evolve.”
Interest in deintensification crystallized nearly 10 years ago as HPV-related head and neck cancers became more common and were recognized as a clear clinical and pathologic subtype. Retrospective studies found a pattern of better outcomes for patients who tested positive for the p16 tumor suppressor protein, which is highly correlated with HPV infection.
Important trials examining this question include the analysis of the phase III Radiation Therapy Oncology Group (RTOG) 0129 trial, published in 2010, which found that patients with locally advanced head and neck cancer that is correlated with HPV had better 3-year rates of overall survival (OS) than patients who were HPV-negative (82.4% vs 57.1%, respectively). Those with HPV-positive disease had a 58% reduction in the risk of death after adjustment for tobacco exposure and other variables (HR, 0.42; 95% CI, 0.27%-0.66%).1
Deintensification had not previously been a focus because patients with OPSCC generally have poor prognoses, according to Saba. Patients with HPV-negative tumors who receive standard treatment have 5-year survival rates of about 50%, he said, and studies often find poorer outcomes for OPSCC subgroups. Research is focused on extending survival, in some cases by testing maximum tolerable dosing, rather than on mitigating the harms resulting from treatment.
However, those long-term effects can substantially affect survivors’ quality of life. That is of particular concern for the growing number of patients who are treated successfully at a relatively young age, many of them nonsmokers with HPV-positive disease, and who experience life-long toxic effects. For example, dysphagia, or swallowing dysfunction, is “a debilitating, depressing, and potentially life-threatening complication” that persists long after treatment is complete.2 The probability of dysphagia increases 19% with every additional 10 Gy of radiation.3
Other late chemoradiation effects can include xerostomia (dry mouth), aspiration of foreign objects, feeding tube dependence, hyperthyroidism, neck fibrosis, and osteonecrosis. “We’re dealing with the patients who are 10 years out from chemoradiation and they’re miserable. And it’s not fixable,” said Robert L. Ferris, MD, PhD, director of the University of Pittsburgh Medical Center Hillman Cancer Center in Pennsylvania.
Rates of HPV-positive oropharyngeal cancers in the United States more than doubled from 1988 to 2004 while the incidence of HPV-negative cancers fell by half.4 The virus is now thought to cause 70% of oropharyngeal cancers in the United States.5
OPSCC has become the most common HPV-associated cancer, overtaking cervical cancer; the category consists of squamous cell cancer types at the base of the tongue, the pharyngeal tonsils, the anterior and posterior tonsillar pillars, the glossotonsillar sulci, the anterior surface of the soft palate and uvula, and the lateral and posterior pharyngeal walls, as well as the anus and rectum.6 The disease predominantly affects men, but from 1999 to 2015, overall rates of OPSCC increased both among both men (2.7%) and women (0.8%) (Figure).6 The shift has been attributed to changing patterns of sexual activity. Oncogenic oral HPV prevalence is higher in men who have more oral sexual partners.7
It’s a little bit of a younger population, and the incidence is rising in developed countries,” said Quynh-Thu Le, MD, a radiation oncologist at Stanford University in Palo Alto, California. “We learned that these cancers behave differently. They tend to be more responsive to radiation and chemotherapy, and their outcomes are better.”
As community oncologists see more patients who are HPV positive, some have been trying to reduce long-term complications by reducing radiation doses or avoiding chemotherapy altogether, Saba said. “That makes us at least some of us feel uncomfortable because people are doing it without clear guidelines. Recent reported results from RTOG-1016 have shown that deintensification by changing the chemotherapy agent is detrimental in terms of overall survival of patients with HPV-related disease. So even though a selected group of patients may qualify for radiation only, not everyone does,” he said.More systematic efforts to test deintensification of chemotherapy support the standard approach, according to clinical trial findings reported in October. Saba said the experience with NRG Oncology/RTOG-1016, a phase III trial that evaluated the systemic therapy that usually accompanies radiation in OPSCC treatment calls for caution.
“We should not deintensify automatically. When we tried to deintensify by removing cisplatin and putting in cetuximab, we failed,” Saba said. “Even though we had so much enthusiasm about deintensification, we should pause and re-evaluate our approached based on RTOG-1016.”
The 805-patient NRG Oncology/RTOG-1016 study was the first randomized clinical trial specifically designed for patients with HPV-positive oropharyngeal cancer. Patients received intensity-modulated radiation therapy (IMRT) at 70 Gy, the standard dose for nonsurgical therapy, and either cisplatin or the EGFR inhibitor cetuximab (Erbitux), with the hope of showing that cetuximab would produce less toxicity while maintaining efficacy.8 Cetuximab is approved for the treatment of patients with locally or regionally advanced head and neck squamous cell carcinoma (HNSCC), a broader class of malignancies that includes OPSCC, in combination with radiation therapy (RT) and in metastatic settings with platinum-based therapy.
In NRG Oncology/RTOG-1016, participants were randomized to receive IMRT for 6 weeks with either 2 cycles of cisplatin (100 mg/m2) every 3 weeks or cetuximab (400 mg/m2 loading dose followed by 250 mg/m2 weekly for 8 doses).
The study was a noninferiority trial designed to determine whether outcomes with the monoclonal antibody with IMRT would come within 5 points of cisplatin plus IMRT. However, the National Cancer Institute (NCI) announced in August that the cetuximab group had worse OS and progression-free survival (PFS), a result that Ferris called “surprising.”
At 5 years posttreatment, patients who received cetuximab (n = 406) plus IMRT had an OS rate of 78% and a PFS rate of 67% compared with 85% and 78%, respectively, for those treated with cisplatin plus IMRT, according to findings reported at the 2018 American Society for Radiation Oncology Annual Meeting. Other outcomes favored the cisplatin regimen (Table).9
“We’ve now established that high-dose cisplatin chemotherapy—in combination with radiation—is the standard of care for HPV-related oral cancers,” Andy M. Trotti, MD, co-lead investigator of NRG Oncology/RTOG-1016 and a radiation oncologist at the Moffitt Cancer Center in Tampa, Florida, said in discussing the results.10
At the same time, cetuximab may be an option for patients who cannot tolerate cisplatin, such as those with severe hearing loss or diabetes-related neuropathy, he noted.
Toxicities were different between the 2 regimens, Trotti said, with renal toxicity, hearing loss, and bone marrow suppression more common with cisplatin and body rash occurring more often with cetuximab.
Overall, the burden of acute toxicities was higher with cisplatin than it was with cetuximab, Trotti said. He developed a T-score system in which acute grade 3/4 AEs observed in the entire study group are divided by the total number of patients. Under that measure, the cetuximab arm had a T-score of 2.35, meaning the average patient in that group had more than 2 high-grade AEs, while patients in the cisplatin cohort had a T-score of 3.19. That translated into a statistically significant 40% increased acute toxicity rate (P <.001).9,10 Under the standard method, the rate of grade 3/4 AEs was 81.7% with cisplatin versus 77.4% with cetuximab(P = .16).
Findings from the De-ESCALaTE HPV trial also suggested that RT with cisplatin was more effective than RT with cetuximab in patients with low-risk, HPV-positive OPSCC, according to results reported at the 2018 European Society for Medical Oncology (ESMO) Congress.11 In all, 334 patients enrolled at centers in the United Kingdom, Ireland, and the Netherlands were randomized to receive RT at 70 Gy in 35 fractions with either cisplatin (3 doses of 100 mg/m2) or cetuximab (400 mg/m2 loading dose followed by weekly 250 mg/m2). The study compared severe AEs, quality-of-life outcomes, OS, and other measures.11
During the 2-year study, there were 10 recurrences and 6 deaths in patients randomized to cisplatin compared with 29 recurrences and 20 deaths in those assigned to cetuximab. Patients in the cisplatin arm had a significantly superior 2-year OS rate than those on cetuximab (97.5% vs 89.4%; HR, 4.99; 95% CI, 1.70-14.67; P = .001). Of note, the trial was not powered for the endpoint of survival; however, the results point in the same direction as the RTOG-1016 findings.
The difference in OS was driven by a difference in locoregional control and distant control. Patients assigned to cetuximab were 3 times more likely to experience a recurrence in 2 years compared with cisplatin, with recurrence rates of 16.1% versus 6.0%, respectively (HR, 3.39; 95% CI, 1.61-7.19; P = .0007).
There was no difference in mean global health status or swallowing scores between the 2 groups. The mean number of overall AEs per patients was not different between the cisplatin and cetuximab arms (5.37 vs 5.45 events per patient, respectively. The incidence of acute or late severe (grade ≥3) AEs per patient (4.81 vs 4.82) or all-grade acute and late toxicity (29.15 vs 30.05 events per patient) was similar between the cisplatin and cetuximab groups, respectively. Significantly more serious AEs (162 vs 95) were reported in the cisplatin arm compared with the cetuximab arm.
“Many patients have been receiving cetuximab with radiotherapy on the assumption that it was as effective as chemotherapy with radiotherapy and caused [fewer adverse] effects but there has been no head-to-head comparison of the 2 treatments,” said lead investigator Hisham Mehanna, MD, chair of head and neck surgery at the Institute of Cancer and Genomic Sciences, University of Birmingham, United Kingdom. “The surprise of the study was in the survival. What we found was significantly worse survival with cetuximab. The number needed to treat to cause harm was 12.”
Taken together, results from the NRG Oncology/RTOG-1016 and De-ESCALaTE HPV trials are being interpreted as definitive findings concerning the role of cetuximab in the broad HPV-positive OPSCC population. One expert noted that more work is needed to determine the impact of molecular alterations, particularly KRAS mutations that are part of the EGFR signaling cascade.
“We now have 2 studies showing that these patients should not be given cetuximab,” Branislav Bystricky, MD, head, Medical and Radiation Oncology Department, University Hospital Trencin, Slovakia, said in a statement issued by ESMO. “Future research should examine whether genotyping for the KRAS variant can select a group of patients that will benefit from cetuximab treatment with radiotherapy.”Another line of research seeks to better identify the patients who would do well with a reduced radiation dose. In EA 1308, an ECOG-ACRIN Cancer Research Group trial published last year, patients with HPV-positive OPSCC were treated with a 3-drug induction regimen of cisplatin, paclitaxel, and cetuximab, and then those with a primary-site complete clinical response (cCR) were given reduced-dose IMRT. The study enrolled 90 patients with a median age of 57 years (range, 35-73) who had HPV- or p16-positive stage III/IV OPSCC.12
After a median follow-up of 35.4 months, the 2-year PFS and OS rates were 80% and 94%, respectively, for evaluable patients with primary-site cCR who were treated with 54 Gy of radiation (n = 51). Patients with lower-volume disease (tumor stage,
At 12 months, significantly fewer patients treated with ≤54 Gy had difficulty swallowing solids (40% vs 89%; P = .011) or impaired nutrition (10% vs 44%; P = .025), respectively, compared with those who had received higher doses. Longterm smokers did not fare as well: Among those treated with the lower radiation dose, 2-year PFS was higher for patients with ≤10 pack-years versus those with >10 pack-years (92% vs 57%; P = .0014), as was 2-year OS (93% vs 86%; P = .040).
As a single-arm trial, EA 1308 was not practice changing, but it showed that chemoselection can be used to identify patients for deintensified radiotherapy and paved the way for future studies, said Ferris, who was a coauthor on the study. “It said there is a subgroup that can get 54 Gy and have 95% to 100% survival. So deintensification can be done safely. The question is, how and for whom?”
A number of studies are attempting to answer that question. A successor trial to RTOG-1016, NRG Oncology’s HN002, has tested the efficacy of lower RT alone in patients with good-prognosis HPV-positive OPSCC, with participants receiving either 60 Gy with cisplatin or 60 Gy with no chemotherapy.13 An NCI steering committee has also approved a follow-up study to HN002 that will provide a similar therapy, while keeping systemic therapy and reducing the RT field sizes as a type of further deintensification.
Another trial, EA3161, will be adding the immunotherapy agent nivolumab (Opdivo) to the backbone of cisplatin and RT, with the focus on a group of HPV-related OPSCCs that have a higher risk for relapse (the intermediate-risk group). The addition of nivolumab represents an intensification of sorts, Ferris said.
Although EA 1308 supports lower-radiation approaches, Ferris noted that it still depended on a relatively rigorous 3-drug combination, which arguably runs counter to the aims of deintensification. “A surgeon would say, ‘I can give you negative margins on an operation and get those lymph nodes out and give you a complete response. Why do we need 3 drugs of chemo over 9 weeks, which is very expensive and very difficult for the patient, when I can give you a complete response surgically in 1 afternoon?’” he said.
Those types of questions show why deintensification is controversial, down to the issue of how to even define it, oncologists said.
“Which group should we deintensify? Do we deintensify in everybody who has an HPV-positive tumor, or do we deintensify in a subset?” Le said. “Should we deintensify the chemotherapy or the radiation? If we deintensify the radiation, should we cut down the dose or the field? Then there are different strategies to deintensify. ECOG-3311 brings in surgery and more radiation: Does that deintensify or not? Then you bring in more chemotherapy up front and reduce the ratiation: Does that deintensify or not? That’s where the controversy is.”
Ferris launched ECOG-3311 in 2013 to evaluate the efficacy of different surgery-first approaches.14 The 4-arm trial enrolled 519 patients with HPV-positive advanced oropharyngeal cancers, including some smokers. They underwent TransOral Robotic Surgery and neck dissection up front and their postoperative pathologic features were used to select subsequent therapy.
The patients with intermediate-risk pathologic factors—up to 4 metastatic lymph nodes or microscopic extracapsular spread—received no chemotherapy and were randomized to either 50 Gy or standard postoperative 60 Gy. Those with low-risk disease, who had up to 1 metastatic node and no extracapsular spread, got the “ultimate deintensification” of no other therapy, which was “very controversial” when the trial launched, Ferris said.
The fourth, high-risk arm included patients with 5 or more metastatic nodes and extensive extracapsular spread. They received the standard 66 Gy and cisplatin, for a total of 3 modalities, which represented an intensification rather than a de-escalation of therapy, Ferris said. ECOG-3311 is expected to report out by next July.
As more patients entered the study, Ferris became increasingly frustrated with the sequence of surgery, chemotherapy, and high-dose radiation that was offered to the high-risk group, he said. “Over the years of enrolling, it began to look more and more Neanderthal to keep blasting these people,” he said. As a next step, he will soon open a trial for that subgroup that begins with surgery but reduces their radiation to just 50 Gy and replaces their chemotherapy with nivolumab, he said.
Ferris’ upcoming trial and the HN002 follow-up study will be among the first testing immunotherapy as a deintensification tool. Other studies are investigating its use more broadly in HNSCC. Nivolumab was approved in 2016 for patients with metastatic or recurrent HNSCC following progression on platinum-based therapy. Last year, Ferris presented data from the open-label CheckMate-358 study showing that nivolumab given a month prior to surgery reduced tumor size in about half of patients with HNSCC, with reductions occurring both in patients with HPV-positive diesease and those with HPV-negative disease.15
“Potentially, those preoperative doses may be a mechanism to deintensify, if somebody was a good responder. Maybe that’s how we pick who can avoid chemotherapy,” he said.
He noted that results from studies have shown that nivolumab may be combined with radiation but there are no data demonstrating the safety of replacing chemotherapy with the PD-1 inhibitor in the treatment of oropharyngeal cancers.
Saba is preparing to launch a trial for patients with locally advanced intermediate-risk HPV-related oropharynx cancer that will provide standard cisplatin RT therapy followed by randomization to either nivolumab maintenance for a year or observation. EA3161 will answer an important question about maintenance therapy for those patients but does not focus on the lower-risk patients, who appear to be the best candidates for deintensification.
“Deintensification, even though it’s happening currently in a lot of HPV-positive disease, is not the overall trend for all HPV-positive cancers. There are cancers that deserve deintensification and cancers that probably do not,” he said.Pembrolizumab (Keytruda) has also been discussed as potentially useful for HPV-associated disease. It is approved for recurrent or metastatic HNSCC that has progressed after chemotherapy based on tumor response and durability rates. Investigators have reported mixed results in trials seeking to establish an OS advantage.
In findings from the phase III KEYNOTE040 trial presented at the 2017 ESMO Congress, pembrolizumab reduced the risk of death by 19% in patients with recurrent or metastatic HNSCC compared with standard-of-care chemotherapy plus cetuximab but did not meet the primary endpoint of a statistically significant improvement in OS.16
In the frontline setting, interim findings from the phase III KEYNOTE-048 trial showed that pembrolizumab as monotherapy or with chemotherapy improved OS and response duration for patients with PD-L1—positive recurrent or metastatic disease compared with the standard regimen of chemotherapy plus cetuximab.17
Lenvatinib (Lenvima) added to pembrolizumab also demonstrated promising activity in patients with HNSCC in an ongoing open-label phase Ib/ II clinical trial, according to data presented at the 2018 American Society of Clinical Oncology Annual Meeting.18 In 22 patients, the objective response rate (ORR) with the combination was 40.9% (9 of 22) and the median PFS was 8.2 months. Separately, tumor mutational burden, PD-L1 expression, and T-cell activated gene expression profile were recently shown to be predictive of response to pembrolizumab in patients HNSCC regardless of HPV status.19
A number of other agents have demonstrated response in head and neck cancers, such as the CTLA-4 inhibitor ipilimumab (Yervoy), Saba said. He mentioned the pan-PI3K inhibitor buparlisib (BKM120), which in combination with paclitaxel reduced the risk of disease progression or death by 35% in patients with advanced HNSCC.20 Combinations of VEGF inhibitors and immunotherapy are also being pursued in several small phase II studies, he said.
The combination of the CDK4/6 inhibitor palbociclib (Ibrance) and cetuximab induced an ORR of 39% in patients with platinum-resistant, HPV-unrelated recurrent and metastatic HNSCC.21