In this study, we demonstrated that while IMRT-based (chemo)radiation can have short- and long-term toxicities, early-stage HPV-associated OPSCC patients experienced excellent survival with acceptable toxicities, low long-term gastrostomy dependency rates and negligible treatment-related mortality (0% in this study). In contrast, prior studies have reported higher toxicity rates, with treatment related deaths of up to 3%2, [i]. Similarly, a frequently cited meta-analysis of the Radiation Therapy Oncology Group treatment intensification trials by Machtay and colleagues reported that 43% of patients treated with primary chemoradiation for OPSCC suffered a severe major late toxicity, including a 10% rate of long-term feeding tube dependency15. This is relevant as these studies are often cited as a reference comparison for primary transoral surgery4,8-11. While TOS can be carried out for advanced stage disease, in most case series, >85% of tumours are limited T1-T2 stage4, 8-10. In contrast, the chemoradiation studies included high rates of advanced T stage disease (>75% in one such study by Ang and colleagues)2, introducing bias into the comparison of historical series. Furthermore, many of these studies were carried out in the pre IMRT era, which would likely impact function and toxicity2, 14. Patients eligible for TOS have low-volume disease that may allow for unilateral radiotherapy and improved normal tissue sparing compared to historical chemoradiation cohorts that include a wider range of patients. There is a deficiency in the literature of reported outcomes of solely early-stage patients. This study reports the outcomes and toxicities of early-stage oropharynx patients undergoing primary radiotherapy with modern techniques, and provides a historical comparison cohort for discussions around surgical outcomes. Rates of toxicity and treatment-related mortality are lower than in previously published cohorts.
Febrile neutropenia is a potentially life threatening and frequent complication of chemoradiation in many studies[ii], [iii]. For example, a study by Bledsoe and colleagues reported a rate of febrile neutropenia of 26% in patients treated with chemoradiation with 2/32 (6.3%) of these patients dying as a direct result15. In contrast, 19.7% of our early-stage patients did not require chemotherapy at all and thus were not at risk of this complication. In the chemoradiation cohort specifically, the rate of febrile neutropenia was only 8.8% and there were no fatalities (Table 2).
One of the strongest predictors of poor patient quality of life following treatment for head and neck cancer is long term gastrostomy tube dependence[iv]. Chemoradiation studies report rates of up to 10% long-term dependence while most TOS studies show lower rates4, [v],[vi]. In a previous systematic review, the majority of studies demonstrated gastrostomy rates of less than 4.5%, with many (6/13) showing rates of 0% suggesting superior swallowing function with primary surgery4. However, this study has 2-year gastrostomy rates of 1%. This is similar to the results of the ORATOR trial, and suggest that swallowing outcomes are similar between the two treatment strategies6. Given the similar survival, treatment selection for early OPSCC should be an informed decision made between clinicians and patients.
In this study, we also attempt to describe potential prognostic indicators for overall and disease-free survival. It is well-established that HPV-associated OPSCC has a significantly better overall and disease-free survival than HPV-negative patients3. In our study, overall survival but not disease-free survival was found to be significantly better in HPV-associated OPSCC, possibly due to limited sample size (Figure 1). In early stage disease, chemoradiation appeared effective regardless of HPV status, and the differences in overall survival may be partially related to the presumed increased incidence of comorbidities in HPV-negative patients and subsequent non-cancer related death. Likely for similar reasons, on multivariate analysis, a history of smoking was associated with a worse overall survival but not disease-free survival (Additional File 1 and 2).
Concurrent chemoradiation with 3 cycles of high-dose cisplatin is currently the standard treatment for locoregionally advanced OPSCC[vii]. However, due to short- and long-term toxicities, and due to the fact that HPV-associated OPSCC is more sensitive to chemotherapy and radiation than HPV-negative OPSCC, there has been much interest in treatment de-escalation[viii], [ix]. Some current strategies under investigation include weekly cisplatin instead of high-dose cisplatin, lower radiation dose, or decreased adjuvant radiation and/or chemotherapy after surgery[x], [xi], [xii]. Other strategies, such as using the EGFR monoclonal antibody cetuximab, have conclusively been shown to provide inferior survival without meaningfully improving quality of life which only further highlights the importance of balancing toxicity with survival.[xiii], [xiv]
There are a number of limitations with this current study. First of all, it is inherently difficult to accurately grade toxicities retrospectively, precluding the inclusion of other common but less severe toxicities such as mucositis, xerostomia or peripheral neuropathy. The follow-up period was limited and thus may have led to an underestimation of long-term toxicities. This data is limited to a single centre and thus may not accurately portray the true variability in the patient population. Lastly, a number of important potential confounders such as socioeconomic factors and compliance with treatment were not addressed.