The majority of the head and neck cancers are oral cancers 1, 2, 3. Oral cancer as defined by the American Joint Committee on Cancer and the Union for International Cancer Control in the tumor-node-metastasis staging classification, includes carcinomas of the oral cavity originating from the mucosal lip, anterior two-thirds of the tongue, buccal mucosa, floor of mouth, hard palate, lower and upper alveolus and gingiva, and the retromolar trigone 4. Oral cancers account for the third highest standardized death rate in countries with low and medium human development index. Globally, Papua New Guinea had the highest age-specific incidence rate for oral cancer with 21.2 and 12 cases per 100,000 population for males and females respectively 5. Oral cancer is ranked the number one cancer among males in Sri Lanka, with an age-standardize incidence rate of 19.1 per 100,000 people in 2019. As elsewhere, in Sri Lanka, it is more common in men than in women, comprising 15% and 3% of all cancers in males and females respectively 6.

There are many different modalities available for treating oral cancers. Surgery, radiotherapy (RT) and chemotherapy alone or in combination are recommended for treating oral cancers 7, 8. RT with or without chemotherapy is used as the primary treatment modality in early stage and un-resectable tumours to avoid anticipated functional and cosmetic defects. RT is also used in patients when the surgery can be high risk due to comorbidities or poor performance capacity to withstand a surgery, in recurrent malignancies and when patient’s preference is RT 9. In Sri Lanka, RT is commonly used in oral cancer patients post surgically. As most of the oral cancer patients are between 50 and 70 years of age, RT is also used alone or with chemotherapy 10.

Although treatments for cancer are employed to improve the patient's health related quality of life (HRQOL), they are associated with several side effects which deteriorate patients’ HRQOL 11. Dermatitis, dysphagia, mucositis, loss of taste, xerostomia, osteoradionecrosis, trismus, candidiasis and radiation caries are some of the commonest side effects 8, 12, 13. There are generic and disease specific tools to assess HRQOL. Disease-specific measures have greater sensitivity than generic measures because they were developed for specific conditions to measure symptoms and impacts associated with that condition 14. The European Organization for the Research and Treatment of Cancer (EORTC) Quality of Life Questionnaire-C30 (EORTC QLQ-C30), assesses details on five functional domains, one global health status domain, three symptom domains and six items. This allows detailed information about how the various domains are influenced by the disease and treatment and overall quality-of-life 10. The EORTC quality-of-life group has developed a site specific tool to measure HRQOL for head and neck cancers (EORTC QLQ-H&N35). The combination of these two questionnaires provide information on the perception of the patient, the impact of the disease and its treatment, side effects of the treatments, impact of medical interventions, as well as its performance concerning various aspects of life 15.

It has been emphasized that the need of prospective studies to obtain insight into the relation between pretreatment HRQOL and outcome after treatment, and the relationship between changes in HRQOL and functioning over the period of time 16. Therefore, it is important to assess the changes of HRQOL due to their current treatment modalities.

A prospective longitudinal study was carried out among oral cancer patients. The patients were recruited from outpatient clinics and hospitalized patients at The National Cancer Institute (Apeksha hospital), Sri Lanka. Biopsy confirmed adult oral cancer patients who were newly diagnosed during the last three months, and were waiting for RT with or without chemotherapy as the initial treatment, and at any stage were included in this study. Patients who had already undergone surgical treatment and patients who receive small doses of radiation therapy as a palliative care were excluded from the study.

After confirming the eligibility criteria, the purpose and the objectives of the study were explained to the patients. The informed written consent was obtained before collecting the data. The short interviewer-administered questionnaire was used to gather socio-demographic data. The clinical information of the patients was obtained from clinical records.

Validated self-administered EORTC QLQ-C30 and EORTC QLQ-H&N35 were used to gather patient-reported outcomes regarding HRQOL at baseline 17. The patients were followed up and the same two questionnaires were completed again by the same patients during last week of RT course and three months after completion of RT.

Ethical approval for the study was obtained from the Ethics Committee, Faculty of Medicine, University of Colombo, Sri Lanka.

At baseline the sample size was 90, during the last week of RT 86 patients, and at three months after RT 75 patients were followed-up. The majority of the sample (68%) was 50 to 69 years of age. Nearly 88% of the sample consisted of males and 91% were married. When considering the site of the oral cancer, 40% had cancer in the anterior 2/3 of the tongue and 22% on the buccal mucosa. Of the sample, 72% were in the late stage of the disease at diagnosis and 63% were treated with chemo and RT. Only 19% of the sample had diseases other than oral cancer (Table 1).

Most of the domain and item median scores of EORTC QLQ-H&N35 were zero at baseline. During the last week of RT all the median scores were more than 50.0 except ‘Teeth’ (33.3, IQR 0.0-66.7), ‘Coughing’ (33.3, IQR 0.0-33.3), ‘Feeding tube’ (0.0. IQR 0.0-100.0) and ‘Weight gain’ (0.0, IQR 0.0-0.0) items. Furthermore, mean scores for all the parameters of EORTC QLQ-H&N35 were higher during the last week of RT compared to baseline. The most pronounced symptom items were ‘Dry mouth’ (89.9, SD 2.2), ‘Sticky saliva’ (89.9, SD 2.4), ‘Pain killers’ (97.7, SD 1.6), ‘Nutritional supplements’ (98.8, SD 1.2) and ‘weight loss’ (94.2, SD 2.5) during last week of RT. Median scores for ‘Speech problem’, ‘Less sexuality’ and ‘Opening mouth’ remained the same from baseline to three months after RT, while other scores were higher. Almost all the domain and item median scores were lower at three months after RT than the last week of RT except for ‘Pain killers’ and ‘Nutritional supplements’, which remained the same (Table 2).

For EORTC QLQ-C30, the median scores for all functional domains were 100.0 and zero for all the symptom domains and single items at baseline. The median score for the ‘Global health status’ domain was 83.3 (IQR 83.3-100.0) at baseline. All the symptom domains and item median values had increased from baseline to the last week of RT, except ‘Nausea and vomiting’ and ‘Appetite loss’, which remained the same. ‘Physical’, ‘Role’, ‘Cognitive’ and ‘Social’ functioning showed more impairment at three months after RT compared to the baseline. Except for the ‘Appetite loss’ item score, all the symptom domain and item scores were lower at three months after RT compared to the last week of RT (Table 3).

The changes in HRQOL assessed by EORTC QLQ-H&N35 and their significance are tabulated in Table 4. The change of median scores for all the domains and items were significant from baseline to last week of RT, from baseline to three months after RT and from last week of RT to three months after RT (p<0.05) except for the ‘Weight gain’ for all three changes and change of ‘Weight loss’ from baseline to three months after RT.

The changes in HRQOL assessed by EORTC QLQ-C30 are presented in Table 5. The changes for all the parameters from baseline to three months after RT were statistically significant except for the ‘Nausea and vomiting’ domain, ‘Dyspnoea’, ‘Constipation’ and ‘Diarrhoea’ items. The changes in HRQOL measured by all the domains and items of EORTC QLQ-C30 were found to be significantly different at the three time points except ‘Diarrhoea’.

To our knowledge, this is the first prospective study conducted in oral cancer patients who received RT with or without chemotherapy to report HRQOL in Sri Lanka. The HRQOL was assessed by using EORTC QLQ-H&N35 and EORTC QLQ-C30 questionnaires and the major findings were as follows. Mean scores of all the domains and items in EORTC QLQ-H&N35 and symptom domains and items in EORTC QLQ-C30 were higher during last week of RT compared to the baseline. This is similar for median values except ‘Feeding tube’ and Weight gain’ items which remained the same. The functional domains of EORTC QLQ-C30 showed the same pattern inversely except for ‘Role functioning’, these scores continuously reduced from baseline to three months after RT. The results were suggestive of deterioration of HRQOL within three months period after RT. Almost all the changes of the HRQOL were statistically significant. The results indicate patients suffer from more symptoms after the RT than before RT even though the preference is that the treatment improves HRQOL 18, 19

A considerable percentage of the sample (17 %) consisted of patients aged more than 70 years. This may be due to the inclusion of the oral cancer patients whose surgical management was not possible due to the old age and therefore, the RT or chemo-RT had been the treatment of choice 19. A prospective study conducted among head and neck cancer patients who receive RT had shown slightly different results in the scores of EORTC QLQ-H&N35. ‘Dry mouth’, ‘Sticky saliva’, ‘Teeth’ and ‘Opening mouth’ have increased three months after RT compared to baseline and one month after RT whereas in the present study, the corresponding scores three months after RT were lower than the last week of RT. In contrast to the present study the baseline symptom scores were also higher. That may be due to treatment variation and variation in the site of cancer as the study had been conducted in head and neck cancer patients and two-thirds of the sample had undergone surgery before RT 20.

Another study reported the mean scores using EORTC QLQ-H&N35 and EORTC QLQ-C30 at baseline, 40 days after initial treatment and one month after RT in much similar manner to the scores of the present study, despite having done the study in head and neck cancer. In that study, 40 days after initial treatment was similar to last week of RT in the present study as the most patients underwent RT for 30-33 days. The HRQOL of head and neck cancer had reduced from baseline to 40 days. The HRQOL had become better than 40 days compared to one month after RT hence it was more deteriorated than the baseline 21.

All the functional and symptom domains and items have changed more than 20 points from baseline mean value to last week of RT in EORTC QLQ-C30 which indicates ‘very much change’ in HRQOL and “Appetite loss” falls to ‘moderate change’ category according to Osoba et al 22. Hence, there is a clinically significant reduction in HRQOL in all the domains and items of EORTC QLQ-C30 according to King, who suggested a change of 10 points to consider as important clinically 23. Furthermore, the deterioration of HRQOL from baseline to last week of RT was statistically significant except “Diarrhoea” and “Weight gain” items.

Of note, the symptoms directly related to the oral cavity were the mostly affected domains and items even after three months of RT. Similarly, ‘Swallowing’, ‘Sticky saliva’, ‘Opening mouth’, ‘Dry mouth’, and ‘Teeth’, showed as highly affected parameters in other studies 21, 24.

In contrast to our study findings, another study results revealed that any of the functional domains in EORTC QLQ-C30 had not changed significantly and only the “Fatigue”, “Pain”, “Insomnia” and “Appetite loss” changed significantly at six weeks compared to baseline. The disparities in the cancer sites, the commonest stage and the RT technique used in two samples, may be the reasons 25. The present study sample had undergone conventional RT (Cobalt 60) and the majority of studies show a reduction in toxicity and better improvement with time than the immediate effect when RT was given using linear accelerators compared to conventional RT 26.

The results of this study will be useful to develop treatment guidelines and the areas that healthcare professionals should additionally focus on when managing patients. Patients who receive RT with or without chemotherapy face more difficulties to cope up with the side effects of RT towards the end of the RT course. Both patients and their caregivers need extra support from health care professionals to overcome these problems 27, 28, 29. There are some limitations of this study. The exact time of assessing baseline HRQOL varied from patient to patient from a few days to just before RT. The tool EORTC QLQ-H&N35 we used, was not specifically designed to assess HRQOL of oral cancer patients. The study was confined only to three months after RT which allows short term evaluation of HRQOL affected by RT with or without chemotherapy. Selection bias cannot be forgotten as some of the oral cancer patients may seek treatments from private sector and other few hospitals where patients are treated with RT. However, high response rate, use of validated commonly used questionnaires, and assess more homogenous group who have not undergone surgery were the strength of this study when finding the HRQOL in oral cancer patients who receive RT.

HRQOL of oral cancer patients declined due to RT with or without chemotherapy from baseline to last week of RT and improved three months after RT than the HRQOL during last week of RT. Nevertheless, it had not come back to the baseline level, even after three months post RT. The changes in HRQOL were statistically and clinically significant from baseline to last week of RT, from baseline and three months after RT, and last week to three months after RT. Our study results showed strong evidence that treating patients with radiotherapy with or without chemotherapy itself has caused distressful, suffering outcomes which change the patients’ quality of life drastically despite curing the cancer. Therefore, when addressing patient management decisions, health care professionals need to pay more attention regarding the perception of the patient.

We thank all the patients who participated in our study, and their families. We are grateful for all the support provided by the staff of Apeksha Hospital, Maharagama, Sri Lanka. We thank Dr. Nelum Samaruthilaka for advice provided on data analysis.