The impact of multimodality integrated positron emission tomography-computed tomography on improving the staging and management of head and neck malignancy: a cross- sectional study

ABSTRACT BACKGROUND: Clinical assessment of head and neck cancers is highly challenging owing to the complexity of regional anatomy and wide range of lesions. The diagnostic evaluation includes detailed physical examination, biopsy and imaging modalities for disease extent and staging. Appropriate imaging is done to enable determination of precise tumor extent and involvement of lymph nodes, and detection of distant metastases and second primary tumors. OBJECTIVE: To evaluate the initial staging discrepancy between conventional contrasted computed tomography (CT) and 18F-fluorodeoxy-D-glucose positron emission tomography/computed tomography (18F-FDG PET/CT) and its impact on management plans for head and neck malignancies. DESIGN AND SETTING: Prospective cross-sectional study in two tertiary-level hospitals. METHODS: This study included 30 patients with primary head and neck malignant tumors who underwent contrasted computed tomography and whole-body 18F-FDG PET/CT assessments. The staging and treatment plans were compared with the incremental information obtained after 18F-FDG PET/CT. RESULTS: 18F-FDG PET/CT was found to raise the stage in 33.3% of the cases and the treatment intent was altered in 43.3% of them, while there was no management change in the remaining 56.7%. 18F-FDG PET/CT had higher sensitivity (96% versus 89.2%) and accuracy (93% versus 86.7%) than conventional contrast-enhanced computed tomography. CONCLUSION: Our study demonstrated that 18F-FDG PET/CT had higher sensitivity and accuracy for detecting head and neck malignancy, in comparison with conventional contrast-enhanced computed tomography. 18F-FDG PET/CT improved the initial staging and substantially impacted the management strategy for head and neck malignancies.

Informed consent was obtained from all the enrolled patients.
The exclusion criteria were that the subjects should not be children, individuals with acute or chronic inflammatory disease, pregnant patients, lactating mothers, terminally ill patients or patients with any previous malignancy. All the patients selected (above 18 years old) were thoroughly examined by otorhinolaryngology surgeons, and biopsies were taken from suspicious regions.
All the patients underwent CECT and whole body 18  The change in management intent and the incremental information obtained after both imaging procedures had been done were compared and analyzed. The percentage of management changes implemented due to discrepancies between the imaging methods was recorded. The clinical impact of PET/CT was considered 'high' if it changed the treatment modality, and 'low' if there was no change in the treatment modality or intent. 7 All the patients were monitored through regular follow-up at a specialist clinic. The cumulative survival rate among the patients was estimated from the date of diagnosis to the date of death due to any cause or the date of the last follow-up. It was noted whether any patients were lost to follow-up or were still alive at the end of the follow-up period. The five-year overall mean survival rate was calculated, and the mean survival time in months according to sociodemographic characteristics, tumor stages and treatment received was also analyzed.

RESULTS
Out of the 30 patients in this study, 60% (18/30) were male and 40% (12/30) were female. According to ethnicity, the majority were Chinese (56.7%; n = 17), followed by Malay (40%; n = 12) and Indian (3.3%; n = 1). The mean age (with standard deviation, SD) was 49.9 years (± 14.5) with first and second peak age incidences in the age ranges of 30-39 years and 60-69 years, respectively. Nasopharyngeal carcinoma was the commonest malignancy (56.7%), and the next commonest was carcinoma of the larynx and malignancy of the oropharynx (10%). A list of the primary sites of tumors is shown in Table 1, while the patients' characteristics, clinical and histopathological diagnosis are presented in Table 2.
All of the 30 patients underwent pre-treatment radiological assessment with CECT and 18 F-FDG PET/CT for the purpose of disease stratification through the AJCC 7 th edition TNM staging, in order to determine the intended management plan. The CECT and 18 F-FDG PET/CT findings were classified as true positive (positive imaging study that was confirmed histopathologically), true negative (normal imaging study with no further evidence of cancer), false positive (positive imaging study with no histopathological evidence of cancer) or false negative (normal imaging study with further proven cancer). 8 Through 18 F-FDG PET/CT, it was found that there were 27 true positive cases, one false positive, one false negative case and one true negative case. Three patients who were suspected of having benign lesions following a conventional clinical assessment were proven to be malignant cases after histopathological examination and 18 F-FDG PET/CT. Two of these patients were initially diagnosed through conventional staging as having a thyroglossal cyst, but malignancy was proven through 18  The treatment plans were altered in the cases of 43.3% (13/30) of our study group patients, while there was no management change in  Table 3.
Among the patients whose staging increased, 30% (3/10) of them benefited from addition of neoadjuvant chemotherapy.
Distant metastases were identified in six patients and the management plans were changed from definitive to palliative intent.
One of our patients with carcinoma oropharynx, for whom oncological surgery with radiotherapy had been planned, was found through 18 F-FDG/PET/CT to be developing lung metastases, which was confirmed through cytological tests. The management plan was therefore changed to palliative therapy. Another patient with metastatic adenocarcinoma of the skull also had a change in the treatment plan to palliative intent. Among the remaining four patients, who had nasopharyngeal carcinoma (NPC), two presented skeletal metastases and the other two were seen to have mediastinal nodal metastases. 18 F-FDG PET/CT decreased the staging of 6.6% of the patients (n = 2). One of these cases consisted of postoperative tonsillar sarcoma, in which there were low-activity lesions in distorted anatomy, which reflected the post-surgical change. The other case was incorrectly diagnosed as parotid lymphoma, and the final histopathological diagnosis was benign oncocytoma (Figure 1).
The influence of 18 F-FDG PET/CT on stage migration and its impact on management intent are shown in  assessment, to 96% and 93% from 89.2% and 86.7%, respectively.
The negative predictive value from the CT assessment was lower than the NPV from PET/CT imaging.
After a period of five years, we performed a search in the patients' records at the specialist clinic. We found that 10 patients  Table 4).
Mean survival time in months according to sociodemographic characteristics, tumor stages and treatment received was analyzed and is presented in Table 5.
For all of our patients with head and neck malignancies, the   results for diagnosing and staging of head and neck squamous cell carcinoma, compared with standard imaging modalities. 12 In our study, we sought to prospectively evaluate the influence of 18 F-FDG PET/CT on the initial staging and its impact on the treatment plan for head and neck cancer patients. The data for this study consisted of information from our own patients, in contrast with the data in other, multicenter studies, which relied on medical records. The majority of our study patients (56.7%) were diagnosed with nasopharyngeal carcinoma, which was in accordance with the epidemiological pattern of head and neck cancers in Malaysia. Nasopharyngeal carcinoma is one of the ten most common cancers in this multiracial Southeast Asian country.
Our results demonstrated that 18   The treatment plans were changed in 43.3% (13/30) of our patients, while no management change was made in the cases of the remaining 56.7%. Our study results showed changes that were similar to what was observed by Veit-Haibach et al. 22 In that study, the accuracies of TNM staging using PET/CT and CT were compared, and it was found that staging based on PET/CT imaging changed the therapy for 42% (13/31) of the patients, compared with therapy based only on CT. 22 In another study by El-Khodary et al., treatment changes were made in the cases of 41.7% of the patients. 9 A variety of changes to treatment were made among our patients. These included addition of chemotherapy or radiotherapy and abandonment of localized surgery and radiotherapy with curative intent, which was replaced by treatment with palliative intent. The aim of chemotherapy was shifted from curative to palliative intent in 20% (6/30) of our patients. These patients were in a group at an advanced stage with presence of distant metastases.
Our data were found to be consistent with the findings of previously published studies. 13,[23][24][25] In the present study, 18 F-FDG PET/CT was found to have improved sensitivity and accuracy for detecting head and neck malignancy, in comparison to conventional CECT. The sensitivity, specificity, PPV, NPV and accuracy of 18 F-FDG PET/CT were reported to be 96%, 50%, 96%, 50% and 93% respectively. This was comparable to the study published by Gordin et al in 2007. 26 During the follow-up of our study group patients, we found that 10 patients had passed away. These patients' treatments were therefore reclassified from having curative to having palliative intent. This notably strengthens the argument that PET/CT has a major incremental impact with regard to identifying high-risk patients who do not benefit from aggressive curative treatment.
In interpreting 18 F-FDG PET/CT imaging, the challenges include physiological uptake of fluorodeoxy-D-glucose (FDG) by normal tissues, false positive results due to inflammation, limited resolution of small lesions and motion artefacts. 27,28 Costeffectiveness is the major consideration in deciding whether to use of 18 F-FDG PET/CT as part of the initial imaging. Its cost needs to be weighed against the benefit of early detection of distant metastases, synchronous primary and resulting interventions. 29 Our study had several limitations. The majority of our study cohort were NPC patients, which might have introduced a working bias. Moreover, our sample consisted of a small number of patients with head and neck cancers at different sites. Because of these limiting factors, the results from our study focused mainly on nasopharyngeal carcinoma and may not have reflected the situation regarding other head and neck cancers. Further prospective studies comprising larger patient cohorts are required in order to ascertain the impact of 18 F-FDG PET/CT on the management of various head and neck malignancies.

CONCLUSION
In conclusion, our study demonstrated that 18 F-FDG PET/CT had higher sensitivity and accuracy for detecting head and neck malignancy than those of conventional CECT. 18 F-FDG PET/CT provides additional information and accurate staging, which assist in planning for adequate treatment and in minimizing treatment-related toxicity and functional impairment. From our study findings, we would advocate for incorporation of 18 F-FDG PET/CT into the initial staging of clinically advanced head and neck malignancy.