The impact of the COVID-19 pandemic on head and neck cancer diagnosis: a single-center study

Head and neck cancers (HNC) constitute the sixth most common malignant neoplasm globally, with approximately 930,000 new cases in 2020. The most common locations are the mouth, pharynx, and larynx [1]. While the overall incidence of HNC is slowly decreasing, the incidence of HPV-dependent oropharyngeal cancers is increasing. The 2-, 5-, and 15-year survival rate in patients with HNC is estimated at 64%, 46%, and 21%, with the worst prognosis for the oropharynx and the lower pharynx [2]. In many cases, the disease is asymptomatic for a long time or with non-specific symptoms such as hoarseness, sore throat, or neck tumors. Therefore, most patients report to a doctor only after the appearance of disease symptoms, and more than half of the cases (62.4%–87.7%) are diagnosed in stage III or IV [3]. Unlike colorectal, lung, breast, and cervical cancer, no consensus on HNC screening has been established yet [4].

In 2020, oncological care experienced many difficulties. The growing number of COVID-19 infections, initially in China and then around the world, was recognized by the World Health Organization as a “Public Health Emergency of International Concern” on January 30, 2020, and as a pandemic on March 11, 2020. Then, the lockdown and numerous changes in the organization of healthcare were introduced. Some hospitals adapted to COVID-19 treatment, and other procedures were limited. In the first months of the pandemic, therapeutic options were limited and included: symptomatic treatment, oxygen therapy, venous thromboembolism prophylaxis and systemic steroids [5]. COVID wards suffered from lack of protective clothing, respirators, and staff. Later FDA approved some of the drugs tested for the treatment of COVID-19, such as remdesivir, baricitinib and tocilizumab. Initially, high hopes were associated with the use of chloroquine and hydroxychloroquine, but they are no longer recommended for treatment of COVID-19 due to a lack of clinical data and possible side effects such as ototoxicity [6,7]. During the lockdown, healthcare systems had limited cancer screening as well as cancer diagnosis and treatment. Due to fear of infection, some patients resigned from visiting the doctor. Cancer departments worldwide had to implement solutions enabling the continuation of care while minimizing the risk of infection for patients and medical staff [8]. Particular challenges concerned the head and neck area due to the necessity of examining and performing procedures on the upper respiratory tract and the resulting exposure to aerosol in exhaled air [9].

In HNC patients, delayed treatment results in more advanced disease and a worse prognosis [10]. The relative odds of N upstaging after 40 days without treatment may be 1.40 [11]. Moreover, every 30 days of delay in surgery increases the risk of death by 4.6% and in oropharynx neoplasms by 29% [12]. It can be expected that delaying diagnosis will have a similar effect on disease severity and prognosis.

The aim of this study was to assess the impact of the COVID-19 pandemic on patients with HNC.

172 patients in the Otolaryngology, Head and Neck Surgery Department of the Wroclaw University Hospital with newly diagnosed HNC were enrolled. Table 1 shows the general characteristics of patients. Each group consisted of 86 patients. Most subjects were male (82.6% in the pre-COVID group vs. 73.3% in the COVID group). The mean age was 63.8 years (63.5 vs. 64.1). The youngest patient was 30, and the oldest was 86. The mean BMI was 25.4 kg/m2 (25.9 vs. 24.8). 4% of patients (3.5 vs. 4.6) were underweight (BMI <18.5), and 32.5% (36 vs. 29.1) were overweight or obese. The mean distance from the patient's residence to our hospital was 25.4 km (25.9 vs. 24.8). Both groups were most often in the range of 0–24 km (pre-COVID group: 34.9%, COVID group: 30.2%).

There was no statistically significant relationship (p>0.05) between general characteristics and the treatment period (before and during the COVID-19 pandemic). Table 2 presents the clinical characteristics of patients treated before and during the pandemic. The number (and percentage) of patients in the subgroups differing in the treatment period, tumor stage in TNM classification, histopathological diagnosis, and tumor location are presented, as well as the results of Pearson's chi-square tests (p).

In the COVID group, fewer cases of T0 compared to the pre-COVID group (3.5% vs. 10.5%) were noted. It is related to the number of unknown primary (CUP) cancer diagnoses (3.5% vs. 10.5%). In addition, during the pandemic, there were fewer T1 (20.9% vs. 26.7%) and more T3 tumors (22.1% vs. 11.6%). The distribution of other T values was similar in both groups. Less than half of the patients with newly diagnosed HNC had no lymph node involvement (43% and 46.5%, respectively). N2 value was more common in the COVID group (33.7% vs. 25.6%), as was the presence of distant metastases (8.1% vs. 3.5%), which is reflected in staging. Diagnoses at stage IVC were more frequent during the pandemic (7% vs. 2.3%). Other stages’ contributions were similar in both groups. Histologically, most diagnoses were squamous cell carcinomas (81.4% and 84.9%, respectively), most often of the G2 type (55.8% and 58.1%). The most common tumor location was the larynx (45.3% vs. 47.7%), followed by the hypopharynx (22.1% vs. 14%). The oropharynx (14.1%) was third in the COVID group and CUP (10.5%) was third in the pre-COVID group.

All differences between groups in clinical characteristics of patients were statistically insignificant (p>0.05). However, it should be noted that regardless of the period, more than half of the cases were in stage IV at the time of diagnosis (53.5% before vs. 59.3% during the pandemic).

Table 3 presents symptom duration before hospital admission, history of smoking, and the type of treatment for which patients were qualified after diagnosis. The duration of symptoms was similar in both groups, the most common being between 0 and 3 months (32.5% in the COVID group vs. 32.6% in the pre-COVID group). During the pandemic, we more frequently diagnosed HNC in patients with no smoking history (40.7% vs. 29.1%). Most patients qualified for radical treatment (65.1% vs. 72.1%). Treatment most often consisted of surgery, radiotherapy, or a combination of both.

The frequency of additional interventions on admission to the hospital are presented in Table 4. Eighteen out of 172 patients reported dyspnea on admission, and 11 were from the COVID group. An emergency tracheotomy was required in 16 cases, and 10 were diagnosed during the pandemic. Moreover, 19 patients required to secure feeding with PEG directly after admission, and 14 of them (73.7%) belonged to the COVID group (p<0.05). The odds of performing PEG in the COVID group compared to the pre-COVID group were more than three times higher (OR=3.15, 95% CI: 1.08 to 9.18, p=0.049).

Figure 1.

The COVID-19 pandemic forced changes in the functioning of many medical facilities, led to changes in healthcare priorities, and significantly impacted early cancer diagnosis and screening. In April 2020, in the US, the number of screening tests performed for breast, colon, prostate, and lung cancer decreased by 85%, 75%, 74%, and 56%, respectively [15]. In many countries, cancer care had to be rearranged in terms of screening programs, medical appointments, imaging tests, hospital admissions, and surgeries. These factors are essential for the prognosis of cancer patients, including HNC. Gazzini et al. revealed the negative impact of the COVID-19 pandemic on the diagnosis and stage of HNC in South Tyrol, the further consequences of which will be observed in the future, especially in terms of cancer stage progression [16]. HNC patients diagnosed at University Hospital of Modena during the first phase of the COVID-19 pandemic had a more advanced stage at presentation compared to patients from the previous year [17]. In Wuhan, China, the treatment of patients with HNC was delayed by almost four months due to the pandemic outbreak [18]. In Sergipe, Brasil, nearly 60% of HNC patients experienced delays in cancer care and, moreover, they often became fearful of becoming infected by SARS-CoV-2 [19]. A meta-analysis performed by de Castro et al. revealed a significantly smaller number of surgeries on HNC patients performed in the COVID-19 era [20]. It is estimated that the decrease in the number and subsequent delay in new cancer diagnoses caused by the COVID-19 pandemic will result in more advanced stages of the disease, additional deaths, and worse survival and quality of life [21]. In the UK, delays in cancer diagnostics may result in increased deaths from breast, colorectal, lung, and esophageal cancer within five years after diagnosis by 9.6%, 16.6%, 5.3%, and 6%, respectively [22]. Many studies confirmed the association between treatment delay and poorer overall survival in HNC [10]. In the study, which analyzed HNC cases in the United States, time to treatment initiation of greater than 46–52 days was associated with worse survival [23]. Another study revealed worse survival in early glottic cancer in cases of delay between diagnosis and treatment over 100 days [24].

Kiong et al. noted a 25% decrease in the number of new diagnoses of HNC during the pandemic [25]. HNC originating from the mucosa diagnosed during the pandemic was larger and more advanced in the T stage than in the corresponding period before the pandemic. No differences were found in skin, thyroid, or salivary gland cancers. In our study, the vast majority of cases were cancers originating from the mucosa; however, we did not conduct a separate analysis of this group of tumors. No difference in the total number of new diagnoses or cancer stages was noted during the analyzed period. The difference in obtained results may be related to longer compared periods of time in the case of our study (13 months vs. 6 weeks). The situation may vary depending on the country, cancer treatment center, and regulations that have been altered during the pandemic. Analyses of large national cancer databases will provide more information on this topic. At the moment, the most recent statistics from the Polish National Cancer Registry are from 2019.

In our study, patients more often required a gastrostomy at admission during the pandemic. The decision on the type and timing of nutritional support implementation is complex, although it is based on objective tools, such as NRS 2002 [14]. If there is a need for enteral feeding for more than 4 weeks, a gastrostomy tube (G-tube) should be introduced. Among our patients, the most common method of G-tube implantation was percutaneous endoscopic gastrostomy (PEG). The explanation of why both groups did not differ in terms of BMI but did when it came to PEG is because the PEG decision is based, inter alia, on NRS 2002, which contains not only BMI score but also the presence of recent weight loss or impaired food intake. In order to directly assess the relationship between the COVID-19 pandemic and dysphagia and malnutrition in newly diagnosed patients, it is necessary to conduct a study comparing the results of objective tools such as NRS 2002, Malnutrition Screening Tool (MST) or Scored Patient-Generated Subjective Global Assessment (Scored PG-SGA) [14,26–27].

In our study, diagnostic process was delayed by a factor related to the coronavirus pandemic in some patients. The most common causes included the patient's fear of being infected in a public place or health care facility, the increased number of telemedicine appointments in primary care, which might lead to misdiagnosis, and finally, patient infection and isolation due to COVID-19. However, symptom duration on admission in both groups was similar. A possible explanation for these results is the low accuracy in assessing the duration of symptoms only based on the medical interview. On the other hand, except for postponing hospitalization of patients with positive RT-PCR tests, the in-hospital diagnostic mechanism was mostly undisturbed. While most elective procedures were suspended for a certain period of time, access to the operating room and diagnostic imaging for oncological patients were not limited in our center. Therefore, fewer scheduled non-oncological procedures could translate into less overload in diagnostic imaging and histopathological laboratories, resulting in shorter waiting times for these procedures. Further research assessing referral, diagnostic, and treatment intervals during pandemics is needed to confirm these assumptions.

Many oncological patients were concerned that the COVID-19 pandemic and the resulting healthcare changes would disrupt their diagnostic and therapeutic processes. However, for some cancers, the risk of delaying treatment is more significant than for COVID-19 infection, especially when safe vaccines are available that are highly effective in preventing the disease and reducing the severity of the disease. In addition, by maintaining a sanitary regime, including wearing masks, disinfecting hands and surfaces, and implementing in-hospital procedures such as mandatory PCR testing before scheduled medical procedures, it will be possible to provide oncological care during the COVID-19 pandemic safely.

Our study had several limitations, including a small sample size and retrospective single-center design. In addition, the pre-pandemic and pandemic periods were divided arbitrarily, although the boundary between them was more fluid in practice. Moreover, varying degrees of public restrictions could be observed during the pandemic, and their direct impact on cancer diagnosis delays is challenging to investigate.

HNCs are a diagnostic and therapeutic challenge. They are often diagnosed in advanced stages, which forces treatment intensification. During the COVID-19 pandemic, patients came to the hospital in worse general condition, and more had dysphagia or malnutrition requiring PEG. This can incur delayed costs of HNC treatment resulting from impaired wound healing, prolonged nutritional treatment, and extended hospitalization. However, the pandemic did not affect the stage of HNC significantly. Regardless of the period, more than half of the cases were diagnosed in stage III or IV of the disease. This study was performed in the first months of the pandemic and does not cover the pandemic's long-term effects on cancer patients. Based on the national cancer database, long-term evaluation should assess if there was a difference in screening rates and the number of patients referred to Fast Track Cancer Pathway (number of issued Oncology Diagnosis and Treatment Cards). Unfortunately, these data are not yet available.