The colleagues at the Princess Margaret Cancer Center in Toronto came to similar conclusions when performing a retrospective analysis of 624 stage III/IV OPC patients treated with radiotherapy alone or concurrent chemoradiotherapy (CCRT) in their institute for whom they had p16 data (Huang et al. 2013). Local control, regional control, and overall survival were all significantly improved in the p16-positive cases versus the p16-negative cases at 3 years (p < 0.001). In addition, they focused their interest on the risk of developing distant metastases rather than survival outcome alone. Not only was the natural course of the distant metastases different, i.e., distant metastases in p16-positive OPC can occur later than what is usually seen in the p16-negative cases, but they also noticed that the occurrence of distant metastases was significantly associated with T4 category disease, the degree of nodal involvement, and, moreover, showing the same relationship with smoking history as observed in RTOG trial 0129, reported by Ang et al. (vide supra). The risk of distant metastases in T1-T3, N0-N2a disease seemed rather low both in heavy and light smokers, but a smoking history of >10 pack-years was important in N2b disease, while patients with N2c and N3 disease were at risk of distant metastases irrespective of smoking history (O’Sullivan et al. 2013). The colleagues from Princess Margaret Hospital also reported on differences in the type of distant metastases and the consequences thereof. In the p16-negative cases, distant failure is mostly seen in the lung, followed in frequency by metastases to bone and liver, and in general, these metastases are considered incurable. In the p16-positive cases, two types of distant metastases can be distinguished, i.e., the so-called disseminating type, occurring in multiple organs and in unusual sites, and the so-called indolent phenotype, which still might be cured with salvage procedures (surgery, chemotherapy, or radiation) (Huang et al. 2013).

Although the prognostic importance of HPV for OPC is irrefutable, the impact of the HPV status on treatment response is less established. Retrospective subanalyses in randomized trials are not conclusive on a specific benefit of one particular treatment over the other in HPV-positive OPC patients (Rischin et al. 2010; Lassen et al. 2010). In the DAHANCA 5 study, the benefit of nimorazole only seemed to be present in the p16-negative cohort, and in the TROG 02.02 phase III trial, there was only a trend favoring the tirapazamine arm for improved locoregional control in p16-negative patients. Because of these rather limited and unconfirmed data, the predictive significance of HPV status in patients with LA-OPC needs further study.

Therefore, it should be concluded that the HPV status does not currently alter the management of patients with LA-OPC. In fact, there are no guidelines for the treatment of HPV-positive OPC, neither in the USA nor in Europe. However, there are some proposed strategies, one of which is the use of induction chemotherapy (which had showed to be more efficacious in HPV-positive OPC than in HPV-negative OPC in a prospective trial (Fakhry et al. 2008) to select those patients who might need less intensive locoregional treatment afterward. This concept was studied in Eastern Cooperative Oncology Group (ECOG) trial 1308, originally presented at ASCO 2014, and more recently updated at ASCO 2015 (Cmelak 2015). ECOG 1308 allowed dose reduction of the intensity-modulated radiation therapy (IMRT: (54 Gy/27fx + cetuximab weekly) in HPV-positive resectable stage III or IV OPC patients if a complete clinical response were obtained to 3 cycles of induction chemotherapy, which consisted of cisplatin 75 mg/m² day 1, paclitaxel 90 mg/m² day 1, 8, and 15, and cetuximab 250 mg/m² day 1, 8, and 15, given at 3-week intervals. Those patients who achieved only a partial response or remained stable received full dose of bioradiotherapy (IMRT 69.3 Gy/33fx + cetuximab). The update reported on the symptom reduction observed in those patients, using the Vanderbilt Head and Neck Symptom Survey version 2 [VHNSS V2] at 6 months and 12 months compared to baseline. Difference in difficulty swallowing solids (35 % vs 100 %) reached statistical significance (p = 0.01). A composite analysis evaluating moderate-to-severe symptoms at 12 months for any of the 3 clusters (difficulty swallowing solids, dry mouth, and taste/smell changes) was 70 % vs 100 %. The conclusion of the investigators was that a 15 Gy dose reduction seemed to be able to meaningfully reduce some late toxicities without compromising efficacy (Cmelak 2015). It should be mentioned that the observation time to make a final conclusion on efficacy is rather short to allow for a definitive conclusion. Another deintensification study that makes use of induction chemotherapy (ICT) is the Quarterback Trial. In this trial, stage III and IV HPVOPC patients receive 3 cycles of ICT with docetaxel, cisplatin, and 5-FU [TPF]; when CR/PR is achieved, then patients are randomized to receive 56 or 70 Gy, and when no response is obtained, patients receive standard CCRT. Another approach is using radiotherapy alone rather than CCRT, as done in the ADEPT trial. In this trial, patients with T1-4a, N+ (ECE+) HPVOPC, and negative margins after transoral robotic surgery are randomized to receive RT alone vs CCRT with cisplatin. Finally, the use of bioradiation [BRT] instead of CCRT is investigated in order to study whether BRT leads to less acute and late toxicity. This latter approach is being studied in three studies around the world: RTOG 1016 in the USA, the DeESCALaTE study in UK, and the TROG 12.01 study in Australia. Most likely, the last three studies will be combined for specific analyses. However, again the efficacy outcome data will take many years. Future clinical trials in LA-OPC should at the very least stratify by HPV status. Ideally, HPV+ and HPV− groups should be evaluated in separate trials.