“The status quo for HPV [human papillomavirus]-associated oropharyngeal squamous cell carcinoma is not sufficient.… Our treatment is effective, but the toxicity associated with it is not tolerable.” And HPV-associated oropharyngeal cancer “is a cancer of relatively younger patients,” said Nishant Agrawal, MD, Professor and Director of Head and Neck Surgical Oncology, University of Chicago, at the 2016 Multidisciplinary Head & Neck Symposium sponsored by the Robert H. Lurie Comprehensive Cancer Center of Northwestern University, Chicago.1
Our patients are asking us to do better and reduce toxicity.— Nishant Agrawal, MD
Tweet this quote
We have seen a significant increase in the incidence of HPV-associated oropharyngeal cancer in relatively younger patients, with the median age of diagnosis in the 50s, even patients in their 30s,” Dr. Agrawal said. “These patients have a long life expectancy—you can expect them to live 30 to 50 years or even longer as life expectancy continues to increase—vs head and neck cancer patients with disease related to smoking, who are diagnosed in their 60s and have maybe another 10 to 20 years to live because of other smoking-related cancers and medical comorbidities.”
The long life expectancy and the possibility of quality of life being compromised by chronic adverse events of current head and neck cancer treatments are reasons for considering deintensification. “Acute toxicity is significant with our current treatments,” Dr. Agrawal said. An estimated 80% of patients will have grade 3 toxicity with current treatments, and up to 25% to 60% will “continue to suffer from grade 3 toxicity.” Adverse events include dermatitis, mucositis, xerostomia, and dysphagia and can be “life-altering” for patients.
According to Sandeep Samant, MD, Chief of Head and Neck Surgery at Northwestern Medicine and symposium Chair, “The goal of deintensification is to reduce late toxicity, while maintaining high survival and normal functioning related to swallowing, saliva, and taste.”2 Dr. Samant served as moderator and was one of the speakers at the keynote session on de-escalating therapy in HPV-positive oropharyngeal cancer.
Two Different Diseases
“Most people involved in treating HPV-positive and HPV-negative oropharyngeal cancers recognize that they are two different diseases,” Dr. Samant stated. That statement was supported by responses to an audience poll that showed 70% answered no to the question, “Do you think HPV and non-HPV cancer should be treated similarly?” and 96% said no to the follow-up question, “Do you think in 5 years’ time, they will be treated the same way?”
The goal of deintensification is to reduce late toxicity, while maintaining high survival and normal functioning related to swallowing, saliva, and taste.— Sandeep Samant, MD
Tweet this quote
“There is probably a reason to believe that patients with HPV are currently being overtreated. There is just no clarity on how that treatment needs to change safely without giving up on survival rates,” Dr. Samant said.
“HPV association has a markedly positive impact on prognosis,” Dr. Agrawal noted. While most patients with head and neck cancer in early stages “generally do very well,” with 5-year survival rates of 70% to 90%, the survival rate for patients with stage III or IV disease “unfortunately plummets” to 30% to 60%, Dr. Agrawal reported, “with the exception of HPV-associated head and neck cancer,” which leads to 5-year survival rates that can exceed 85%.
Two papers have supported this, Dr. Agrawal noted. An evaluation of 96 patients with stage III or IV head and neck cancer3 “shows that at a median follow-up of approximately 40 months, patients with HPV-positive oropharyngeal cancers have a 2-year overall survival of 94% vs those with HPV-negative disease, who have a 2-year overall survival of 58%.” A retrospective analysis4 of 323 patients with advanced oropharyngeal squamous cell cancer enrolled in a randomized clinical trial found a 3-year overall survival rate of 82.4% for HPV-associated disease vs 57.1% for HPV-negative disease, and progression-free survival of 73.7% vs 43.4%, Dr. Agrawal reported.
Predictors of Dysphagia
Quality-of-life scores are generally better for patients who are HPV-positive than for p16-negative patients, and quality of life “generally improves to near baseline in 12 months,” Dr. Agrawal said. “Unfortunately, a fair number of patients still suffer from chronic dysphagia. Even at 2 years after radiation therapy, 15% of patients had grade 2 swallowing dysfunction and 8% had progressive dysphagia, so their swallowing is going to continue to get worse.” Patients may also have chronic xerostomia. “The dry mouth improves,” Dr. Agrawal said, “but it never gets back to 100%.”
Dr. Samant reviewed several studies of factors associated with dysphagia. A study of 122 patients who had definitive radiation for locally advanced head and neck cancer “found a dysphagia rate of 38.5%, and the single most important predictor was concurrent administration of chemotherapy,”5 Dr. Samant noted. In a study from the Netherlands6 with 529 head and neck cancer patients, “the most important predictors of grade > 2 dysphagia were concurrent chemoradiation, accelerated fractionation, or bilateral neck radiation,” he stated. Another study from the Netherlands found that the probability of dysphagia is dependent on radiation dose to the superior and middle constrictor muscles.7 After 55 Gy, “the probability of dysphagia kept increasing by almost 20% for each 10-Gy additional dose,” Dr. Samant said.
Depending on the study, anywhere from 5% to 85% of patients with dysphagia will require percutaneous endoscopic gastrostomy (PEG) tube placement, Dr. Agrawal reported, and “some of these patients are PEG-dependent for the rest of their long lives.” These patients require special food preparation, compensatory mechanisms, or food limitations. “You are asking a 50-year old to change his or her lifestyle for the next 30 to 40 years,” he stated.
“Interventions are important,” Dr. Agrawal stressed. They include early swallowing therapy; speech pathology before, during, and after treatment; and preventive exercises. However, “the key is deintensification,” he maintained.
Deintensification/De-escalation Approaches
Dr. Agrawal said that he was “very surprised” by the results of a study finding that patients will accept a marginally lower survival to avoid the toxicity associated with severe treatment. In the study, 81% of patients indicated they preferred reduced treatment intensity,8 Dr. Agrawal noted. “Our patients are asking us to do better and reduce toxicity.”
Dr. Agrawal listed several approaches to deintensifying treatment and reducing toxicity:
- Reduce the number of treatment modalities. “We want patients to have single-modality therapy or, at most, double-modality,” he said.
- Reduce the intensity/dose of a given modality without compromising efficacy. For example, radiation doses can be de-escalated for patients who respond favorably to induction chemotherapy. Dr. Agrawal suggested omitting chemotherapy for patients at low risk of metastatic disease and treating these patients with radiation or surgery alone.
- Try transoral robotic approaches to surgery vs traditional open surgical approaches. “At most academic institutions, transoral robotic surgery has been transformative in the care of patients, and the patients do well,” Dr. Agrawal said. “You change a 7-day hospitalization for 12-hour surgery to essentially an outpatient procedure, where patients stay in the hospital for 23 hours, swallow on the same day or on postoperative day 1, and by 2 weeks, have pretty good function and recovery.” Transoral surgery also provides accurate pathologic staging to treat the patient appropriately, he added.
“The role of immunotherapy is yet to be defined. Right now, we only think about it in the recurrent/metastatic setting, but you can also consider it with curative intent, adding it to transoral robotic surgery, radiation, or chemotherapy,” Dr. Agrawal said. He predicted that “immunotherapy is going to be a game-changer in the field, just as transoral robotic surgery has been,” but emphasized that at this point, immunotherapy should not be used outside of a protocol or clinical trial in the primary setting.
Role of Induction Therapy
“Not all patients with HPV-related cancers do well,” Daniel Haraf, MD, Professor, Department of Radiation and Cellular Oncology, University of Chicago Medicine, reminded the symposium participants.9 He described several studies aimed at determining whether distant disease is an issue in worsening prognosis and whether induction therapy works in this setting.
An analysis by Brockstein and colleagues looked at five trials from 1989 to 1998 involving a total of 337 patients with locoregionally advanced head and neck cancer.10 Results were compared for patients treated with either type I protocols (intensive induction chemotherapy, a two-drug concomitant regimen, and once-a-day irradiation) or type II protocols (no induction therapy, a three-drug concomitant regimen, and twice-a-day irradiation).
[Induction chemotherapy] reduced distant metastases [in patients with oropharyngeal cancer], but it didn’t improve survival.— Daniel Haraf, MD
Tweet this quote
“Survival was no different for type I and type II protocols,” Dr. Haraf reported. At 5 years, overall survival was 46% for patients receiving the type I protocol vs 48% for those on the type II protocol. The corresponding progression-free survival rates were 59% and 62%. “But with the type I protocols, local failure predominated,” Dr. Haraf said. “With the type II protocols, where you got rid of induction, you saw a reversal of the patterns of failure.”
The 5-year local failure rate was 31% for type I vs 17% for type II protocols. Distant failure rates were 13% for type I vs 22% for type II protocols. “Local failure was driven by T stage; distant failure was driven by N stage, especially N2b or N3 disease,” Dr. Haraf stated.
“Data suggested distant failure was important,” Dr. Haraf said. It is possible that distant failure may have been underestimated because of rapid local failures among patients receiving induction and low-dose concomitant chemotherapy, he added.
Reduced Distant Metastases
The phase III DeCIDE trial in head and neck cancer found no difference in survival between patients randomly assigned to receive chemoradiation with or without induction chemotherapy.11 The study involved “the highest-risk patients,” all with N2 or N3 disease, Dr. Haraf noted.
At 3 years, overall survival was 72.8% for patients receiving chemoradiation vs 74.9% for those also receiving induction chemotherapy. The corresponding progression-free survival was 58.6% vs 67.4%. “Local control was identical,” Dr. Haraf stated. “Distant recurrence was significantly different (P = .04). Distant failure was decreased when we gave induction chemotherapy.”
The study was considered a negative trial, Dr. Haraf noted. “It reduced distant metastases, but it didn’t improve survival.” He pointed out that the study was closed early and remained underpowered for the survival endpoint, and there were not enough events. “It cannot be assumed that induction chemotherapy doesn’t work, because a positive trial may prove benefit, but a negative trial does not prove the opposite,” Dr. Haraf said.
Limiting Radiation Fields
A trial of response-adapted volume de-escalation used induction chemotherapy with cisplatin, paclitaxel, and cetuximab (Erbitux) to address distant metastases and regional micrometastases as well as to predict response and curability.12 Every other week, patients with measurable locally advanced head and neck squamous cell cancer who achieved complete or partial responses (≥ 50% reduction in the sum of tumor diameters) after two cycles of induction chemotherapy, received paclitaxel, fluorouracil, hydroxyurea, and 1.5-Gy twice-daily radiotherapy to gross disease only, to a total dose of 75 Gy. Patients with stable disease (< 50% response, “nonresponders”) received the same combination with 75 Gy to gross disease and 45 Gy to the first echelon of uninvolved nodes.
At 3 years, overall survival was 79.1% for responders vs 83.1% for nonresponders, and the corresponding progression-free survival was 79.7% vs 72.0%. Local tumor control at 3 years was 94.4% for responders vs 79.8% for nonresponders, and the corresponding distant tumor control rates were 94.3% vs 89.6%.
“The majority of local failures (12/13, or 92.3%) were in-field, and most of them (11/12, or 91.7%) were in the highest-dose area,” Dr. Haraf reported. “On multivariate analysis including tumor and treatment characteristics, only good response to induction chemotherapy predicted for decreased G-tube dependency on treatment and at 3 months and 6 months posttreatment,” he added.
OPTIMA is an ongoing trial (though closed to accrual) at the University of Chicago stratifying therapy among 61 patients with locoregionally advanced, HPV-positive head and neck cancer to minimize adverse events. Patients receive three cycles of induction chemotherapy with carboplatin on day 1 and nab-paclitaxel (Abraxane) on days 1, 8, and 16. Following radiologic assessment of response, low-risk patients with responses ≥ 50% receive low-dose radiation therapy, whereas those with responses of 30% to 50% receive low-dose chemoradiotherapy, and those with responses of < 30% receive standard chemoradiotherapy. For high-risk patients, those with responses ≥ 50% receive low-dose chemoradiotherapy, and those with responses of < 50% receive standard chemoradiotherapy.
“Induction is still relevant,” Dr. Haraf asserted. It can be used to address distant disease, to decrease the size of radiation fields (and thereby reduce toxicity), and to predict how to reduce doses in HPV-associated cancers, since “not all HPV cases are cured.”
Need to ‘Respect This Cancer’
Among patients with HPV-associated oropharyngeal cancer, “the cure rates are high, 80% to 90%, but not 100%, and you have to respect this cancer,” Robert Haddad, MD, noted.13 Dr. Haddad is Disease Center Leader, Head and Neck Oncology Program, Dana-Farber Cancer Institute and Harvard Medical School, Boston.
“Now that we are 15 years into this HPV epidemic, we see a lot of distant failures with HPV-positive patients,” he said. “Young men, 45 to 50 years old, with neck masses that are HPV-positive, small primaries, big lumps, those are the typical patients.”
“HPV-positive and HPV-negative oropharyngeal cancers are two different diseases, but they both need to be respected, and both need to be treated aggressively with radiation, chemotherapy, and surgery,” Dr. Haddad said. “In a large number of trials that looked at the HPV cohort, you consistently see better outcomes. These patients have higher survival rates when you are comparing them to the HPV-unrelated group, but again, you are not curing everybody. Also, in some of these studies, the follow-up has been short, only 2 to 3 years. We don’t have 5-year or 10-year follow-up data.”
Bolus Cisplatin Every Few Weeks
Currently, there are “multiple standards of care—radiotherapy alone, chemotherapy/radiotherapy, different chemotherapies; you can do surgery and then radiotherapy, or surgery and then nothing…. The bottom line is that when you use chemoradiotherapy, the standard of care is bolus cisplatin every few weeks with radiation.” While weekly regimens have been used, “there are no phase III data to tell us that weekly cisplatin is as good as bolus cisplatin,” Dr. Haddad asserted.
HPV-positive and HPV-negative oropharyngeal cancers are two different diseases, but they both need to be respected, and both need to be treated aggressively with radiation, chemotherapy, and surgery.— Robert Haddad, MD
Tweet this quote
The Eastern Cooperative Oncology Group (ECOG) 130814 trial among patients with HPV-associated resectable squamous cell carcinoma of the oropharynx “was interesting and important because it did use an induction strategy. Patients got induction chemotherapy first to tease out the complete responders. And those complete responders went on to receive deintensification,” Dr. Haddad explained. Patients with clinical complete responses had low-dose intensity-modulated radiation therapy plus weekly cetuximab, whereas those with partial responses or stable disease got full-dose intensity-modulated radiotherapy plus weekly cetuximab.
“From this study, we learned there were some groups not appropriate for deintensification—T4 patients, N2c patients, smokers. Those patients did not do as well as many other patients, and they are currently excluded from trials looking at deintensification,” Dr. Haddad noted. At 2 years, the progression-free survival rate among patients receiving low-dose intensity-modulated radiotherapy was 80% and overall survival was 93%, but for smokers with more than a 10 pack-year history, progression-free survival was 57% and overall survival was 86%. The best results with low-dose intensity-modulated radiotherapy was in smokers with fewer than 10 pack-years and non-T4, non-N2c disease who had 2-year progression-free and overall survival rates of 96%.
“This approach remains investigational,” Dr. Haddad noted. “Further studies are planned.”
Two New Trials
Radiation Therapy Oncology Group (RTOG) 1016 is a large trial with almost 1,000 patients, looking at the question, “Can we use cetuximab, an epidermal growth factor receptor inhibitor, instead of bolus cisplatin in patients with head and neck cancer?” Dr. Haddad explained. Although the trial is complete, results are not yet available.
The ECOG E3311 trial for HPV-positive patients is “using surgery first to select patients who might receive a lower dose of radiation or chemoradiation,” Dr. Haddad said. “The trial is dropping chemotherapy altogether, taking those patients who appear to be at good risk—with low T and N classifications—and randomizing these patients to low- or standard-dose radiotherapy alone or standard-dose radiotherapy with weekly cisplatin or carboplatin. This is a phase II trial, so it is not likely to be practice-changing, but it might allow us to design a bigger trial.”
With the results of these two trials, “we will have some indication of how patients with HPV oropharyngeal low-volume, low-risk disease can be treated. But we do need mature follow-up of these trials, not just 2 or 3 years, because many treatment failures appear later on,” Dr. Haddad said. “We now have patients we have followed for 10 or 15 years, and we are seeing those failures.”
Dr. Haddad commented, “A number of factors aside from HPV might be relevant to the deintensification discussion.” In addition, testing for p16, the viral subtype in most patients with HPV-associated disease, “might not be sufficient to enroll a patient on a deintensification trial.”
Challenging the Guidelines
According to the National Comprehensive Cancer Network® (NCCN®) guidelines, “we should not be de-escalating therapy,” Dr. Samant pointed out. For advanced oropharyngeal cancer, the guidelines recommend advanced chemoradiation, 70 Gy with bolus cisplatin every 3 weeks, for 2 to 3 cycles, with options for weekly cisplatin. The guidelines do acknowledge, “In general, the use of concurrent chemoradiation carries a high toxicity burden.”
There is a “300% increase in acute toxicity with the addition of chemotherapy,” Dr. Samant said, and three RTOG trials (91-11, 97-03, 99-14) using standard chemoradiation or a variety of different regimens showed a 43% incidence of severe late toxicity.15 In that analysis, the curve for late toxicity “keeps rising until it flattens out at 36 months,” Dr. Samant said. “Did it flatten out because toxicity stopped occurring, or did it flatten out because we don’t have the data? I have seen patients 14 years out from concurrent chemoradiation with late toxicity.”
In an interview with The ASCO Post, Dr. Samant said, “We know the long-term toxicity of treating cancer of the head and neck is not so much from the surgery, particularly if the surgery is performed in a conservative transoral manner. It is from the morbidity of intensive radiation, which is made even more potent by adding chemotherapy. So if you can remove the tumor through a function-conserving transoral technique and remove the lymph nodes, you might avoid second-line therapy.”
Role of Primary Surgery
“With primary surgery, if you resect the tumor with negative margins, you get a pathologic stratification of risk, so that you may be able to select patients who will not get any further treatment or may get a reduced postoperative dose of radiation, restricting the use of concurrent radiation to only those patients who truly need it,” said Dr. Samant.
At the University of Tennessee, Memphis, experience with 43 patients with stage III/IV oropharyngeal cancer, 76% of whom were p16-positive and 56%, smokers, 16% got no additional treatment following surgery and node dissection. “The tumor was removed by operating in clean planes with just a thin margin,” Dr. Samant reported. At a median follow-up of 40 months, “42 of those patients are alive; 1 person died. The predominant mode of failure was distant metastasis…. There were no locoregional failures.”
Overall, studies with patients who had surgery as part of their treatment had higher overall and progression-free survival rates. “We are seeing a pattern here,” Dr. Samant said. “If you adjust for smoking and HPV status, you might come out ahead if you resect the tumor.”
Functional Outcomes
Several studies have compared the functional outcomes of transoral robotic surgery or other primary surgery to chemoradiation. A study comparing 20 consecutive patients undergoing transoral robotic surgery to 20 consecutive chemoradiotherapy recipients found swallowing function returns more quickly and is closer to normal with primary surgery than with chemoradiation.16
“Primary surgery produces comparable survival and possibly improved disease control outcome compared to standard chemoradiation,” Dr. Samant concluded. If the primary is small and amenable to transoral resection, the surgical deintensification benefits are favorable adverse-effect profile, quick surgical recovery, and rapid, immediate control of all known disease, as well as more accurate stratification of therapy. Functional outcomes are also better, he added. ■
Disclosure: Drs. Agrawal, Samant, and Haraf reported no potential conflicts of interest.
References
