Facial pain, health-related quality of life and trismus-related symptoms up to 5 years post-radiotherapy for head and neck cancer

Head and neck cancer (HNC) is the fifth fastest increasing cancer type in Sweden 1 and accounts for over 900,000 new cases annually worldwide 2. Approximately half of the patients with HNC present with late tumor stages, i.e., T3 or T4 3. Treatment strategies for HNC include surgery, radiotherapy (RT) and chemotherapy. In early-stage HNC, unimodal therapy is common, whilst multimodal approaches are usually indicated in more advanced stages 4.

Pain is a common symptom among HNC patients, either due to the cancer itself or as a result of acute or late adverse effects of treatment [5‐7. Cancer-related pain after treatment has been associated with negative effects on quality of life 8. Furthermore, trismus is another burdensome complication that may occur in up to 40% of patients with HNC after oncological treatment [9‐11. Trismus, defined as a maximal interincisal opening (MIO) of ≤ 35 mm 12, has been associated with pain, decreased health-related quality of life (HRQL), as well as eating difficulties and maintaining oral hygiene 13, 14. Radiation fibrosis of the masticatory muscles is thought to be the primary cause of postradiation trismus 15, 16.

Few studies have to date described radiation-induced trismus and its association to pain longitudinally. The longest thus far is by this research group, for up to 3 years after completion of radiotherapy in a cohort half the size of the current study 17, 18. The study found facial pain to be reported by over half of the HNC cohort before oncological treatment and at 3 years post radiotherapy. Patients reporting facial pain at the 3-year follow-up had a lower MIO than the patients that did not suffer from facial pain. Other studies have reported on oral pain specifically, with cross-sectional or retrospective study designs 19, 20 and hence, studies investigating long-term prevalence of pain in patients with HNC after treatment are scarce [21‐23.

As the prognosis of HNC is improving with modern oncological treatments and a large subgroup of HNC are expected to live many years after their cancer treatment, research focusing on mapping pain and improving HRQL after HNC treatment is warranted. Hence, the aim of this prospective observational study is to investigate the prevalence of facial pain, as well as its effect on HRQL and trismus-specific symptoms in patients with HNC for 5 years post-RT. A secondary aim includes investigating the correlation between facial pain and the degree of maximal interincisal opening.

This longitudinal observational cohort study describes pain and HRQL in 194 patients with HNC at multiple time points after completion of radiotherapy with curative intent and is to our knowledge the longest prospective study to date examining the association between facial pain and restricted mouth opening in this setting.

Study results show that a large proportion of treated HNC patients still suffer from facial pain long-term. Approximately 50% of the study population reported some degree of facial pain at baseline before starting oncologic treatment, which then increased to 70% at 3 months post treatment completion. This decreased to 54% at 12 months post-RT completion and finally dropped to 41% at 5 years post-RT. These results are similar to previous findings from other studies. In a cross-sectional study by Ren et al. 23, a HNC cohort of 505 survivors with a mean survivorship of 4.6 years and a history of cancer treatment of radiotherapy (81.5%), surgery (62.9%) and chemotherapy (58.1%) filled out surveys asking if they had had pain in the last week. If yes, they were asked to grade the pain on a numerical pain scale from 0 (no pain) to 10 (as severe as it could be). They were given a body map to point out where in the body they experienced the pain. 45% of HNC survivors with a median follow-up of 3 years post cancer treatment reported pain. The locations of pain were mainly reported being in the “neck and throat”, followed by “head and oral cavity” and lastly “shoulder”. In another study by Cramer et al. 21 from 2018, patients were asked to self-report grade of pain, from none to severe followed by a secondary assessment of pain as measured by the University of Washington quality of life questionnaire. The received cancer treatment of the cohort was mostly multi-modal including surgery, RT and chemotherapy alone or in combination, where 83% had received RT and 60% surgery. 45.1% of 175 HNC patients with a median of 6.6 years after diagnosis reported pain. These patients were however included from a survivorship clinic which may theoretically result in an overrepresentation of reported pain in that specific group. Furthermore, type of cancer pain is not specified in the study. A systematic review by Epstein et al. 31 reported that 50% present with orofacial pain at initiation of treatment (chemotherapy and radiation for HNC) and the number increases to 81% during therapy, dropping to 70% at the end of therapy with 36% of patients still presenting with orofacial pain at 6 months post-therapy. Pain was often assessed using validated quality of life questionnaires such as EORT QLQ-C30, EORTC QLQ-HN35, University of Washington Quality of Life questionnaire (UW-QOL), as well as visual analog scale and study specific questionnaires like Oral Mucositis Daily Questionnaire among others.

According to the results from the EORTC QLQ-C30 and HN35 in our study, the pain group also reported worse HRQL on most scales and items in both questionnaires. Hence, a clear association was seen between reports of presence of facial pain and negatively affected HRQL. The majority of other post treatment studies regarding HNC, pain and quality of life include pain in assessment of more broadly evaluated quality of life as also noted in the mentioned systematic review by Epstein et al. 29. Furthermore, prospective follow-up studies for years on the subject of pain after oncological treatment are difficult to find. This further complicates comparison of results. In the above mentioned the cross-sectional Cramer et al. study on patients with HNC following treatment, 46% of the patients reporting pain reported low overall quality of life versus 12% of those who reported to be pain-free 21. Pain was assessed in two formats, first by asking about presence and degree of pain, then pain was secondarily assessed in the UW-QOL survey.

Furthermore, patients who scored higher (worse) regarding pain, also appear to have a decreased MIO. Albeit this correlation was weak during the first year post-RT, a moderate correlation was shown towards the end of the study period. A possible explanation for this change in degree of correlation strength may be that acute treatment toxicities which are not directly linked to mouth opening, including oral mucositis, cause pain and discomfort during the first time-period post-RT 32. Factors supporting this theory may also be found in the patient-reported HRQL, where scores were worse for symptoms such as Fatigue (EORTC QLQ-C30), Senses and Social eating (EORTC QLQ-HN35) at 12 months post-RT compared to 5 years post-RT in the group reporting pain. A similar pattern was observed for the EORTC QLQ-HN35 single item Dry mouth, again potentially hinting at consequences of acute treatment toxicity.

Albeit caution should be advised when drawing conclusions regarding the causative effects of a multifactorial phenomenon such as the perception of pain, our results may point to the possibility of trismus causing pain or vice versa. In an interventional, matched control study by Andréll et al. 18, HNC patients with post-radiation trismus underwent a jaw exercise program (n = 50) and were compared to a matched control group who did not undergo the structured jaw exercise program (n = 50). Fifty-nine percent of the subjects reported facial pain before oncologic treatment and more than half (51%) of the HNC survivors with trismus reported facial pain at three years post treatment completion. The patients with HNC in the Andréll study all had post-RT trismus. The higher prevalence of pain in that study population indicates that there is an important relationship between a lower MIO among HNC survivors and the risk of reporting pain. Another important finding in the Andréll study that underlines the relationship between facial pain and trismus was that the patients undergoing the structured jaw exercise program had 9 times higher chance of being pain-free at 3 years following treatment. Nonetheless, there are numerous other potential causes of head and neck pain after oncologic treatment for HNC, including osteradionecrosis, lymphedema, burning mouth syndrome and neurological damage 33.

The strength of this study lies in its prospective design with a long follow-up period of 60 months. Furthermore, the objective measurement of MIO, the use of validated, well used patient reported outcome measures and the size of the cohort also contribute to its strength. The study may be limited by the fact that the patients who dropped out of the study had somewhat worse overall health at baseline. Recurrent HNC is an important cause of local pain and the patients in the study who developed a recurrent disease were excluded. Hence this potential cause of reported pain was minimized, possibly further potentiating the risk of survival bias. Furthermore, there might be a slight risk of recall bias considering the recall time of the primary variable in this study being 1 month. In the more recent version of GTQ, the recall time for pain has been changed to 1 week. There is also limited information regarding the use of specific pain medication or if other methods have been used for alleviating pain, as well as the exact location and cause of the facial pain.

Finally, most of the patients had received chemoradiotherapy, and no direct comparison of pain outcomes was made between those who received chemoradiotherapy and those who received radiotherapy alone. Therefore, the effect of chemotherapy versus radiotherapy alone could not be distinguished in this study.

The study findings demonstrate that facial pain was common among HNC patients before oncological treatment as well as up to 5 years post-RT. Reductions in MIO were associated with significantly more facial pain, particularly in the long term and patients who reported facial pain also reported worse HRQL compared to those reporting no pain. The degree of facial pain needs to be evaluated continuously by clinicians to ensure that other possible methods or medications for alleviations are offered to increase the HNC patients’ chance of optimal well-being.