Cross-sectional study of 1-to-5- year head and neck cancer survivors

Head and neck cancer (HNC) accounted estimated 41,000 cases in Brazil in 2020, corresponding to 7.9% of all neoplasms, except for non-melanoma skin cancer.1 The treatment of HNC is largely based on tumour site and disease stage, and mostly involves surgical resection associated with radiotherapy (RT) and/or chemotherapy.2 Despite the radiotherapy technology advances, the toxic effects of anticancer treatment remain challenging, potentially leading to early and late side effects, such as mucositis, oral opportunistic infections, dysgeusia, ageusia, trismus, dysphagia, odynophagia, chronic pain, xerostomia (XA), dental hypersensitivity, radiation-induced tooth decay, and osteoradionecrosis.3 4 Patients usually report a worsening of quality of life (QoL) due to these side effects.

The overall survival of HNC patients has improved over the last decades, but still depends on tumor location.5 Recent data from South America show three-year survival rates of 56.0% for larynx cancer, 54.7% for oral cavity cancer, 48.0% for oropharynx cancer, and 37.8% for hypopharynx cancer.6 In Brazil, survival comparison between two time periods (2001-2006 and 2007-2012) showed stable results for oral squamous cell carcinoma (SCC) but a significant increase for oropharyngeal SCC.7 Consequently, a large cohort of HNC survivors with complex needs is identified, highlighting the importance of studies on this population, focusing on QoL issues,8 and on the development of cancer survivorship initiatives. Although the theme of the HNC profile has already been addressed previously, the inclusion of a complete oral examination is usually lacking.9 Thus, the present study aimed to determine the demographic and clinical profile of HNC survivors, and oral features that could affect QoL.

Patients (adults ≥ 18 years old) of both sexes, with complete remission of HNC after 1-5 years of antineoplastic therapy including RT were included in this study.

This was an observational, cross-sectional study of 184 HNC patients treated with RT through a medical records review. Demographic and clinical data included gender, race, marital status, place of origin, level of education, smoking history, alcohol consumption, comorbidities, topographic location of the tumour, time from first medical evaluation to diagnosis, and the therapy combined with RT. All patients were contacted and invited to attend the cancer outpatient of a university medical service, for an interview on QoL and clinical examination for the number of healthy teeth, XA, and salivary flow, but only 40 patients agreed to participate. This study was approved by the institutional review board of the institutions (CAAE 27765714.0.000.5419).

QoL was assessed using the oral health impact profile (OHIP-14) questionnaire, which had been validated and translated into Portuguese language.10 The OHIP-14 measures the impact of oral health on QoL; it is composed of 14 questions covering seven domains: functional limitation, physical pain, psychological discomfort, physical disability, psychological disability, social disability, and disability of oral health condition.11 12 The maximum score of OHIP-14 is 70, which would indicate a low QoL. XA was assessed using the xerostomia inventory (XI), previously translated into Portuguese and validated by Da Matta et al. (2012).10 The instrument is an 11-item rating scale that covers both experiential and behavioural aspects of XA. The total score can range from 11 to 55 and a high score indicates extremely dry mouth.10 13-15 Salivary flow was assessed by determining the stimulated whole salivary flow rate (SWSFR). Subjects were asked to chew a small piece of silicone for five minutes, and then spit the saliva produced into a test tube. The volume was measured in millimeter, and flow rates <0.70mL/min were considered hyposalivation.4 14 16 Number of healthy teeth (NT) was determined in relation to lost/extracted teeth.

Demographic and clinical characteristics were statistically analysed using the one-sample t-test (p<0.05). Data from the OHIP-14, XI, NT and SWSFR were analyzed by descriptive statistics. Interactions between the clinical variables were assessed by the Spearman's correlation test, considering r values as 0.00-.19 “very weak”, 0.20-.39 “weak”, 0.40-.59 “moderate”, 0.60-.79 “strong”, and 0.80-1.0 “very strong”. The confidence level for all tests was 95%. Statistical analyses were performed using the GraphPad Prism^® software, version 6.0.

From the 184 patients retrospectively evaluated, 145 (78.8%) were male, mostly in the age group of 51 to 60 years (n=71; 38.6%), with a median age of 57.7 (range 14-96) years; 146 (79.3%) were white, and 70 (38.0%) were married. Most of the patients (n=77; 41.8%) were from cities with populations over 100 thousand inhabitants, and 116 (63.0%) had (completed or some) primary education (Table 1).

More than half of the individuals had smoked for more than 30 years (58.7%), in contrast to 13.6% (n=25) who had never smoked (13.6%). Most smokers (37.5% of the HNC patients) consumed from 11 to 30 cigarettes per day, followed by 17.4% who consumed less than 10 cigarettes per day and 16.8% who consumed more than 30 cigarettes per day. The number of individuals who consumed alcohol (n=63, 34.2%) for more than 30 years was significantly higher than the number of non-consumers (n=29, 15.8%) and individuals who consumed for 16-30 years (n=31, 16.8%). Regarding the type of alcoholic beverage, most consumers reported their preference for distilled beverages (n=46, 25%), followed by both distilled and fermented beverages (n=40, 21.7%) and fermented beverages (n=25, 13.6%) (Table 1).

Regarding comorbidities, 46.7% of the patients did not have any type of comorbidities, 30.4% reported arterial hypertension, 8.7% diabetes mellitus, 4.9% dyslipidemia, 3.8% chronic obstructive pulmonary disease, 4.9% heart disease, 3.8% kidney disease, 6% hypothyroidism, and 19.6% other comorbidities.

Considering the tumour site, most (59.2%) of HNC patients had pharyngeal neoplasms, 40.8% of them in the oropharynx, 8.7% in the nasopharynx and 8.1% in the hypopharynx (Table 2). Oral cavity neoplasms represented 22.8% of total neoplasms; 7.6% were located in the floor of the mouth, 1.1% in the hard palate, 1.1% on the lateral tongue, 7.6% in other (non-specified) areas of the tongue with, and 2.7% in the retromolar area; 1.1% of total neoplasms were in the lower lip. Of salivary gland neoplasms, 6.0% were located in the parotid gland, and 0.54% in the submandibular gland; 9.24% of total neoplasms were in facial sinuses (Table 2).

Analysis of the time between patient's first evaluation at this medical service and the diagnosis revealed that 57.6% of the individuals were diagnosed within six months, 21.2% were diagnosed in 7-12 months, and 14.7% in more than 12 months. Only 7.6% of patients were treated with RT alone; most patients were treated with RT combined with chemotherapy (40.22%), followed by chemotherapy and surgery (35.3%), and surgery (16.8%) (Table 2).

Table 3 shows the survival rate profile of the patients in this study, with the largest distribution of individuals being between 0 and 1 and between 1 and 2 in sequence, representing 36% and 30% respectively. For the other years, the distribution was 15% for those between 2 and 3 years, 11% between 3 and 4 years, and finally 8% between 4 and 5 years. The average survival time of these patients was 1.8 years.

Table 4 shows descriptive statistics of the OHIP-14, XI, SWSF and NT results. The mean OHIP-14 score was 24±14, and the most frequent domains that negatively impacted on QoL were functional limitations, physical limitations and pain, which are related to eating problems. The mean XI score was 39±7, and patients had high scores on seven of the 11 items of the questionnaire, which contributed to the high degree of XA. The objective evaluation of saliva flow showed a SWSFR rate of 0.2mL/ min ± 0.25 (ranging from 0.0 to 0.9mL/min), with 37.5% (n=15) of patients showing a saliva flow of 0.0mL/min.

Clinical and radiographic (panoramic when necessary) examination of the teeth showed a mean of 28.5 extracted or lost teeth, and 28 patients were edentulous. The Spearman test showed a moderate, significant positive, correlation between QoL and XA (p=0.0001, r=0.59) and a very weak, significant, negative correlation between SF and QoL (p=0.029, r=-0.35). The correlation between the other variables analysed was not statistically significant (Table 5).

This study described demographic and clinical characteristics of HNC survivors in a university medical service, and showed that most of these patients were men, older than 50 years, with low education level. This is in agreement with results of previous studies conducted in different regions of the world.17-21 The most common tumour locations were the oropharynx (40.8%), followed by nasopharynx (8.7%) and hypopharynx (8.2%). Additionally, we performed clinical examination of 40 patients from this cohort and found that the majority had a mild impact on QoL despite a high level of XA, low SWSFR and high prevalence of edentulism.

The high prevalence of HNC in male patients is reported worldwide, and this type of cancer represents 7.6% of all cancers in Brazil.1 The increased risk of HNC in male patients can be explained by the abuse of substances such as alcohol and tobacco,22-24 which is more common among men than women, in our cohort, 70.6% of the patients had a history of smoking and 71.7% of alcohol consumption. In addition, the most prevalent comorbidity in these patients was arterial hypertension (30.4%), as observed in previous studies,25 26 and can be explained by the fact that alcohol and tobacco consumption are risk factors for this condition.

The XI is a useful tool to determine the severity of XA through scores assigned to issues related to daily activities.27 In the current study, we found an average score of 40 (out of a maximum of 55) resulted from high scores (ranging from 3.6 to 4.7) in seven of the 11 questions 5. It indicates severe XA, which may have been caused by RT. Measurement of the SWSFR was used to evaluate salivary gland function and showed that all patients had hyposalivation (0.2mL/min). Although there is no consensus on SWSFR classification, values of 0.5-0.9mL/min for SWSFR, and below 0.1mL/min for unstimulated whole salivary flow are considered hyposalivation.14 Our findings were similar to other studies,16 28-31 and reinforce the effects of RT on salivary gland function.

A negative impact of RT on oral health related QoL of survivors of HNC patients has been previously reported.31 In our study, QoL was assessed using the OHIP-14, with an average score of 24 (out of a maximum of 70). However, we found a moderate positive correlation (r=0.59) between OHIP-14 and XI scores, and a weak inverse correlation (r=-0.35) between OHIP-14 and SWSFR. Of note, despite the high score in the XI, low SWSFR and the high number of edentulous patients, we observed a low OHIP-14 score, suggesting a mild impact on QoL. Considering the 1-to-5-year period, patients experienced hyposalivation, it seems reasonable to suggest that a process of adaptation might have occurred. Soldera et al. (2020)31 evaluated the QoL of HNC patients three months after RT and identified that hyposalivation had a high impact on QoL. Then, XA can become mild over time even with low SF.16

The main strength of this work was the characterization of both demographic and clinical profile of HNC survivors with complete remission after antineoplastic treatment. However, there are some limitations, including the lack of information in the medical records, and the subjectivity of the OHIP-14 to assess QoL, which may be affected by individual and cultural issues. Another point to be discussed is the absence of data regarding the quality of life of these same patients prior to HNC and treatment, preventing this baseline from being compared to the patient's current condition. This missing baseline also makes it difficult to compare the impact of surgery on the patient's salivation because salivary flow was not measured before treatment, making it difficult to identify its impact on hyposalivation or xerostomia. Another important fact is the lack of knowledge of the technique and radiation dose that the patients received during the radiotherapy, making it difficult to identify its influence on each patient. Another limitation of this study was the small number of patients subjected to clinical oral examination (40/184; 21.7%), which limited more robust statistical analyses to better explore the correlations.

The fact that most patients were older than 60 years, had a low income, were users of the public health care, and lived outside the region of the study may have affected the willingness of patients to participate in the study.

It was possible to observe that the demographic and clinical profile of our group of HNC survivors with a 1-to-5-year of complete remission was similar to those reported worldwide: predominantly white males, 51 to 60 years old, consumers of alcohol and/or tobacco, and who have some comorbidity. Most of the neoplasms were pharyngeal, taking up to 6 months to be identified. When they were diagnosed, most patients were treated with a combination of radiotherapy and chemotherapy, which had the effect of XA in 37.5% of the individuals. It could also be seen that low SWSFR had a negative impact on QoL when XA was present.