Abstract
To evaluate the role of 18FDG-WBPET-CT, Examination under anesthesia (EUA), and multiple-site biopsy in detecting the occult site in head & neck carcinoma of unknown primary (HN-CUP). In this prospective study, 22 patients with diagnosed CUP, after a thorough outpatient endoscopic evaluation of upper airway and radiological evaluation (CT/MRI) that ruled out a primary lesion were included. These patients subsequently underwent whole-body PET-CT and EUA. Based on the presence of suspicious findings ( +) or their absence (−) on 18FDG-WBPET-CT (P) and EUA (E), we divided the patients into 5 groups: P−E−, P−E + , P + E−, P + E + , and P + or E + . All these patients underwent bilateral palatine tonsillectomy, bilateral nasopharyngeal biopsy, and ipsilateral lingual tonsillectomy for identification of occult primary. Out of 22 patients, the primary could be detected in 4 patients (18%) after the workup (three in the oropharynx and one in the hypopharynx, all ipsilateral). 18FDG-PET-CT suspected primaries in 7 patients; biopsy was positive for three (sensitivity-75%, specificity-77%, PPV-43%, NPV-93%). Out of 5 patients, who had suspicious findings on EUA, 3 of the biopsies revealed malignancy (sensitivity-75%, specificity-88%, PPV-60%, NPV-94%). Both PET-CT and EUA when combined, yield a NPV of 100% if both are negative and PPV of 100% when both are positive for suspicious findings. No primary was identified in the absence of a suspicion by PET-CT or EUA. Without a suspicion on 18FDG-WBPET-CT and EUA, there is a limited role of multiple-site biopsies in patients of HN-CUP.
Keywords: Carcinoma of unknown primary, Occult primary of head and neck, Metastatic cervical lymphadenopathy, Positron emission tomography, Tonsillectomy
Introduction
Carcinoma of unknown primary (CUP) is defined as a metastatic lymph node without a detectable primary lesion at the time of presentation. This definition is a dynamic one, and detection of a primary site with clinical, radiological, or pathological workup leads to auto-exclusion of the patient from this definition [1–3]. The literature cites the incidence of CUP as 2–10% of head and neck cancers, and oropharynx is supposed to be the most common site for an occult primary [4, 5]. The most common histological subtype detected in a metastatic lymph node is squamous cell carcinoma (SCC) [6]. Detection of primary has been associated with an improved survival among the patients of CUP [4]. Hence, the patient usually undergoes an exhaustive clinical, radiological, surgical and pathological workup to locate the occult primary lesion, following which if no primary lesion is found, the patient is classified as true CUP [5]. We conducted this study on patients whose primary could not be detected after an initial clinical and radiological evaluation to ascertain the role of 18FDG-WBPET-CT, examination under anesthesia (EUA) and multiple-mucosal biopsies to identify the sites of occult primary.
Methodology
This prospective study was conducted at our tertiary care center in North India from June 2016 to April 2018 after clearance by the institutional ethics committee (IECPG/408/6/2016). We included adult patients presenting to our outpatient setting with histologically or cytologically proven cervical lymph node metastasis (level I–V) and no evidence of primary lesion anywhere in the head and neck based on 7th AJCC staging.
Initial Evaluation
The patient was evaluated in the outpatient setting using fibreoptic evaluation of the upper aerodigestive tract along with base of tongue (BOT) palpation. Contrast-enhanced computed tomography (CECT) from the skull base till the thoracic inlet was performed for all the patients. All the sites of the head and neck were evaluated by both endoscopy and radiology and were reported as normal, suspicious, and likely malignant. A Magnetic Resonance Imaging (MRI) was done if prompted by a suspicious finding on CECT. Patients with < 18 years of age, who refused to participate, had a poor performance status, who had severe claustrophobia, were pregnant, had any history of anaphylactoid reactions with 18FDG or iodine contrast agent, or had evidence of a definite primary or distant metastasis on radiology were excluded.
18FDG-WBPET-CT Protocol
Included patients underwent 18FDG-WBPET-CT as per the standard protocol and an experienced nuclear medicine physician [SAS] evaluated the findings independently. He was blinded to the clinical and structural imaging findings. PET-CT images were evaluated for an area of increased FDG uptake and were reported as normal and suspicious for primary. Depending upon the location, any positive spot on the whole-body scan was considered to be either the primary tumor or a metastasis.
Examination Under Anesthesia (EUA) and Biopsy Protocol
Patients underwent nasopharyngoscopy, direct laryngoscopy, and hypopharyngoscopy under general anesthesia. The evaluating surgeon was blinded to the findings of 18FDG-WBPET-CT during EUA. Once EUA was completed, the patient underwent bilateral palatine tonsillectomy, ipsilateral lingual tonsillectomy, bilateral nasopharyngeal biopsy, and in addition PET/EUA guided biopsies were performed, histopathological examination (HPE) was done using the standard technique of serial step sectioning along with appropriate immunohistochemistry (cytokeratin, p40) to identify the probable site of primary tumor.
Statistical Analysis
The results of the patients whose primary could be detected after this exhaustive workup were used to report sensitivity, specificity, PPV and NPV of 18FDG-WBPET-CT and EUA. Based on the presence of suspicious findings ( +) or their absence (−) on 18FDG-WBPET-CT (P) and EUA (E), we divided the patients into P−E−, P−E + , P + E−, P + E + , and P + or E + . We then compared the histopathology results of each category and hence calculated the positivity rate in each subgroup. Statistical analysis was conducted using SPSS® version 24.0.
Results
The STARD flowchart (Fig. 1) depicts the patient recruitment and the investigative work-up each patient went through to detect the occult primary lesion. The clinico-demographic characteristics of all the included 22 patients have been summarized in Table 1. 18FDG-PET-CT showed suspicious findings in seven patients, and subsequently directed biopsy was positive in three patients (Sensitivity-75%, specificity-77%, PPV-43%, NPV-93%). No patient had a reported ‘likely malignant’ lesion. It failed to detect one suspicious lesion in the right pyriform sinus of size 5 mm, which was identified on EUA (Fig. 2). Out of the 5 patients who had suspicious findings on EUA (including the lesion in pyriform sinus), 3 of the biopsies were positive for malignancy (sensitivity-75%, specificity-8%, PPV-60%, NPV-94%). EUA failed to pick up a submucosal lesion which was reported as suspicious on 18FDG-WBPET-CT (Fig. 3). The detailed data of radiology and EUA is tabulated in Table 2.
Fig. 1.
STARD flow diagram depicting the sequence of patient recruitment, their evaluation with 18FDG-WBPET-CT and EUA and the histopathology results
Table 1.
The clinico-demograhic characteristics of the 22 included patients of this study
| Characteristics | Number of patients, n(%) |
|---|---|
| Gender | |
| Male | 20 (90%) |
| female | 2 (9%) |
| Age (in years) | |
| Mean ± SD | 55.8 ± 8.89 |
| Addictions | |
| Smoking | 22 (100%) |
| Alcoholism | 2 (9%) |
| Levels of lymph node | |
| Level II alone | 9 (41%) |
| Level II + others | 12 (54.5%) |
| Level IB | 1 (4.5%) |
| Cytopathology | |
| Well Differentiated SCC | 10 (45.4%) |
| Poorly differentiated carcinoma | 12 (54.5%) |
| Neck status (All initially T0, M0) | |
| N1 | 2 (9%) |
| N2a | 5 (22.7%) |
| N2b | 2 (9%) |
| N2c | 1 (4.5%) |
| N3 | 12 (54.5%) |
| Suspicious findings | |
| 18FDG-WBPET-CT | 7 (32%) |
| EUA | 5 (22%) |
| PET-CT and EUA both | 2 (9%) |
SCC = Squamous cell carcinoma
FDG-WBPET-CT = Fluorodeoxyglucose-whole body-positron emission tomography
EUA = Examination under anesthesia
Fig. 2.
EUA showing a 5 × 5 mm ulceroproliferative growth in right pyriform sinus (a), which was missed on axial contrast-enhanced CT (b) and on axial cuts of fused 18FDG-WBPET-CT (c), as shown by the arrow
Fig. 3.
Clinical image shows no lesion on oropharygeal examination (a) whereas in PET a suspicious uptake was noted in right palate and palatine tonsil (b, arrow) which on histopathology revealed malignancy
Table 2.
The radiological and EUA details of the patients with a suspicious finding in PET-CT, EUA or both
| S. no | Level of lymph node | 18FDG-PET-CT | EUA findings | Final histopathology |
|---|---|---|---|---|
| 1 | Right level II, III, IV | Suspicious right tonsillar fossa | No suspicious lesion | No evidence of malignancy |
| 2 | Left level II | Suspicious left fossa of Rosenmuller | No suspicious lesion | No evidence of malignancy |
| 3 | Right level II | No suspicious lesion | Right PFS ulcer 5 mm size | Severe dysplasia in pyriform sinus (treated as malignancy) |
| 4 | Right level II and III | No suspicious lesion | Unhealthy mucosa in right maxillary sinus | No evidence of malignancy |
| 5 | Left level II | No suspicious lesion | Left posterior pharyngeal wall irregularity | No evidence of malignancy |
| 6 | Left level II, III, IV, V | Suspicious left tonsillar fossa | Left tonsil induration | Left tonsil squamous cell carcinoma |
| 7 | Left level II, III, V | Suspicious left palatine tonsil | Irregular mucosa in left vallecula | Left vallecula squamous cell carcinoma |
| 8 | Right level II | Suspicious right palatine tonsil | No suspicious lesion | Right tonsil squamous cell carcinoma |
| 9 | Right level IB | Suspicious right fossa of Rosenmuller | No suspicious lesion | No evidence of malignancy |
| 10 | Right level I | Suspicious right lateral pharyngeal wall | No suspicious lesion | No evidence of malignancy |
Based on the above findings, we formed four groups of the patient populations (Table 3). As shown in the table, the P-E- group did not reveal any primary, nor did random biopsies in any other group. Thus, the P−E− group yielded an NPV of 100%. The P + E + group had 2 patients, and we were able to locate the occult primary at the suspicious area in both the cases, yielding a positive predictive value of 100%.
Table 3.
Patient distribution in the four sub-groups
| Subgroup | Total patients | Number of patients with primary detected | Positive predictive value | Negative predictive value |
|---|---|---|---|---|
| P−E− | 12 | 0 | – | 100% |
| P + E− | 5 | 1 | 20% | – |
| P−E + | 3 | 1 | 33% | – |
| P + E + | 2 | 2 | 100% | – |
| P + or E + | 10 | 4 | 40% | – |
P−E−= No suspicious findings on PET and EUA
P + E− = Suspicious findings on PET but not on EUA
P−E + = No suspicious findings on PET but present on EUA
P + E + = Positive for suspicious findings on PET and EUA both
P + or E + = Positive for suspicious findings on PET or EUA
Discussion
The current understanding prompts us to evaluate for an occult primary when faced with a metastatic lymph node. The causation of CUP has 2 primary theories: (i) Presence of a microcarcinoma which due to some genetic alterations has a greater tendency to metastasize regionally and distally and leaves the primary site dormant. Multiple genetic mechanisms have been proposed by various authors, including chromosomal aberrations, aneuploidies, translocations, and point mutations [3]. (ii) Another speculation is immune-mediated spontaneous regression of the primary. In either case, genetics plays an important role and the scientific community lacks specific data in this regard.
The most commonly involved lymph node was the level II group, which primarily drains the pharyngeal axis. The pharyngeal axis is supposed to be the core of all unknown primaries as its anatomical structure, especially the tonsillar crypts; precludes a detailed evaluation and has a potential to harbor occult primaries. In agreement with our data, the western literature too cites the oropharynx as the most common site for occult primary. Most of the western authors attribute oropharyngeal occult carcinomas to Human Papilloma Virus (HPV) infection [7–10]. Twelve of our 22 patients had poorly differentiated or non-keratinizing carcinoma on lymph nodal FNAC, however, in the absence of p16 marker study on their lymph nodal tissue, it is difficult to determine the role of HPV in causation. A previous study from our center has determined the prevalence of HPV infection in oropharyngeal carcinomas to be about 23% only [11]. As all our patients were smokers, we would like to attribute the occult primaries in our study to microcarcinomas caused due to smoking rather than HPV.
Cross-sectional imaging in the form of CECT and CEMRI is essential to detect an occult tumor (usually submucosal) which might have been missed on clinical evaluation. However, these modalities lack sensitivity and often need to be supplemented by nuclear imaging in the form of whole-body PET-CT. CUP has been cited as a leading indication of 18FDG-WBPET-CT in the international nuclear medicine guidelines [12]. A meta-analysis by a Zhu et al. reported a very favourable sensitivity of 0.97 (95% confidence interval [CI] = 0.63–0.99), detection rate of 0.44 (95% confidence interval [CI] = 0.31–0.58), but a low specificity of 0.68 (95% CI 5 0.49–0.83) in 18FDG-PET-CT for cervical metastases of unknown source [13]. Studies by Mani et al., Rusthoven et al. and Karapolat et al. [14–16] too have reported high sensitivity and negative predictive value of WBPET-CT. In our study, there was no definite reported malignancy on WBPET-CT; but it had 75% sensitivity and 93% NPV making it an important diagnostic tool before moving to definite histopathological workup.
Neither could we detect any occult carcinoma in P−E− group (12 patients), nor could we detect any malignancy in the specimens with no prior suspicious finding on 18FDG-WBPET-CT or EUA in other subgroups. This shows that random biopsies as well as tonsillectomy seem to have a limited role unless provoked by a finding in 18FDG-WBPET-CT or EUA, as was shown in a similar study by Tanzler et al. [17]. There are, however, some contradicting opinions. McQuone et al. [18] suggest that bilateral tonsillectomy should be done as it does not increase the morbidity of the surgery. Koch et al. [19] reported a 25% incidence of an occult primary in the contralateral tonsil. Nagel et al. [20] reported a substantial increase in the detection of occult primaries when lingual tonsillectomy and palatine tonsillectomy were combined. We, however see no merit in obtaining these tissue specimens in the absence of a suggestion by the preceding evaluation.
Detection of the primary site has been reported to improve survival and decrease treatment morbidity by limiting the required volume of the tissue to be irradiated. We could detect the primary site in four out of 22 patients in our study, out of which 3 had an occult primary in the oropharynx and one in the right pyriform sinus. These patients received targeted IMRT with concurrent chemotherapy. The remaining patients were classified as true CUP and subsequently underwent non-targeted treatment in the form of panmucosal radiotherapy (35#/70 Gy) with chemotherapy, increasing the treatment morbidity. In a contradicting study, Mistry et al. have reported comparable survival of true CUP patients from the patients whose primary was known, when they treated such cases with curative intent with surgery and panmucosal radiotherapy [21].
There are few studies from the subcontinent on CUP and its diagnostic algorithm. This study proposes a preliminary work-up that can be followed in the management of CUP, in geographical areas where the prevalence of HPV and Ebstein-Barr Virus (EBV) is low. We have included all consecutive patients of CUP barring the ones that had to be excluded. These are a few merits of our study.
Limitations
As our study started prospectively in 2016, we did not evaluate for HPV or EBV status in the metastatic lymph nodes of our cohort, as this recommendation was not in place till 2018 when the AJCC released the 8th edition guidelines. We did not compare the survival outcomes of the patients in whom the primary was detected and the patients with true CUP as our intent of the study was to know only the incidence of primary detection. Although 46 patients with clinico-radiologically CUP presented to us initially; only 50% could be included in our study because of advanced disease at presentation leading to a small sample size.
Conclusion
Our cohort of CUP chiefly harbored oropharyngeal microcarcinomas, i.e. three out of four detected carcinomas. Although in isolation the PET-CT and EUA had a high NPV only, together they had a high NPV and PPV (100%), hence we argue in favor of employing both the modalities in the workup for a CUP case. We did not find any role of multiple-site mucosal biopsies (tonsillectomy, lingual tonsillectomy, and nasopharyngeal biopsy) unless provoked by a suspicious finding on 18FDG-WBPET-CT and EUA.
Acknowledgements
We would like to thank Prem Sagar, Shuchita Singh, Smriti Panda, Shikharani Patel, Darwin Kaushal, Sarthak Tripathy and Bhinyaram Jat for their help.
Funding
None.
Compliance with Ethical Standards
Conflicts of interest
The authors declare that they have no financial or non-financial conflicts of interest.
Ethical Approval
The study was approved by the institutional ethics committee (ref. ID – IECPG/408/6/2016). All procedures performed in studies involving human participants were in accordance with the ethical standards of the institution and with the 1964 Helsinki declaration and its later amendments or comparable ethical standards.
Informed Consent
An informed written consent was obtained from all participating patients included in the study.
Footnotes
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
References
- 1.Pavlidis N, Fizazi K. Carcinoma of unknown primary (CUP) Crit Rev Oncol Hematol. 2009;69:271–278. doi: 10.1016/j.critrevonc.2008.09.005. [DOI] [PubMed] [Google Scholar]
- 2.Jereczek-Fossa BA, Jassem J, Orecchia R. Cervical lymph node metastases of squamous cell carcinoma from an unknown primary. Cancer Treat Rev Neth. 2004;30:153–164. doi: 10.1016/j.ctrv.2003.10.001. [DOI] [PubMed] [Google Scholar]
- 3.Stella G, Senetta R, Cassenti A, Ronco M, Cassoni P. Cancers of unknown primary origin: current perspectives and future therapeutic strategies. J Transl Med. 2012 doi: 10.1186/1479-5876-10-12. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Harper CS, Mendenhall WM, Parsons JT, Stringer SP, Cassisi NJ, Million RR. Cancer in neck nodes with unknown primary site: Role of mucosal radiotherapy. Head Neck. 1990;12(6):463. doi: 10.1002/hed.2880120603. [DOI] [PubMed] [Google Scholar]
- 5.Chernock RD, Lewis JS. Approach to metastatic carcinoma of unknown primary in the head and neck squamous cell carcinoma and beyond. Head neck Pathol. 2015;91:6–15. doi: 10.1007/s12105-015-0616-2. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Strojan P, Langendijk JA, Corry J, Woolgar JA, Rinaldo A, ad hominem Frcse, et al. Contemporary management of lymph node metastases from an unknown primary to the neck: II. A Review of therapeutic options. [DOI] [PubMed]
- 7.Boscolo-Rizzo P, Schroeder L, Romeo S, Pawlita M. The prevalence of human papillomavirus in squamous cell carcinoma of unknown primary site metastatic to neck lymph nodes: a systematic review. Clin Exp Metastasis Neth. 2015;32:835–845. doi: 10.1007/s10585-015-9744-z. [DOI] [PubMed] [Google Scholar]
- 8.Fotopoulos G, Pavlidis N. The role of human papilloma virus and p16 in occult primary of the head and neck: a comprehensive review of the literature. Oral Oncol England. 2015;51:119–123. doi: 10.1016/j.oraloncology.2014.10.018. [DOI] [PubMed] [Google Scholar]
- 9.Ren J, Yang W, Su J, Ren X, Fazelzad R, Albert T, et al. Human papillomavirus and p16 immunostaining, prevalence and prognosis of squamous carcinoma of unknown primary in the head and neck region. Int J cancer United States. 2019;145:1465–1474. doi: 10.1002/ijc.32164. [DOI] [PubMed] [Google Scholar]
- 10.Mehanna H, Beech T, Nicholson T, El-Hariry I, McConkey C, Paleri V, et al. Prevalence of human papillomavirus in oropharyngeal and nonoropharyngeal head and neck cancer–systematic review and meta-analysis of trends by time and region. Head Neck United States. 2013;35:747–755. doi: 10.1002/hed.22015. [DOI] [PubMed] [Google Scholar]
- 11.Bahl A, Kumar P, Dar L, Mohanti BK, Sharma A, Thakar A, et al. Prevalence and trends of human papillomavirus in oropharyngeal cancer in a predominantly north Indian population. Head Neck United States. 2014;36:505–510. doi: 10.1002/hed.23317. [DOI] [PubMed] [Google Scholar]
- 12.Boellaard R, O’Doherty MJ, Weber WA, Mottaghy FM, Lonsdale MN, Stroobants SG, et al. FDG PET and PET/CT: EANM procedure guidelines for tumour PET imaging: version 1. 0. Eur J Nucl Med Mol Imaging. 2009;37(1):181. doi: 10.1007/s00259-009-1297-4. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Zhu L, Wang N. 18F-fluorodeoxyglucose positron emission tomography-computed tomography as a diagnostic tool in patients with cervical nodal metastases of unknown primary site: a meta-analysis. Surg Oncol. 2013;22(3):190–194. doi: 10.1016/j.suronc.2013.06.002. [DOI] [PubMed] [Google Scholar]
- 14.Mani N, George MM, Nash L, Anwar B, Homer JJ. Role of 18-fludeoxyglucose positron emission tomography-computed tomography and subsequent panendoscopy in head and neck squamous cell carcinoma of unknown primary. Laryngoscope United States. 2016;126:1354–1358. doi: 10.1002/lary.25783. [DOI] [PubMed] [Google Scholar]
- 15.Rusthoven KE, Koshy M, Paulino AC. The role of fluorodeoxyglucose positron emission tomography in cervical lymph node metastases from an unknown primary tumor. Cancer United States. 2004;101:2641–2649. doi: 10.1002/cncr.20687. [DOI] [PubMed] [Google Scholar]
- 16.Karapolat I, Kumanlıoğlu K. Impact of FDG PET/CT for the detection of unknown primary tumours in patients with cervical lymph node metastases. Mol Imaging Radionucl Ther. 2012;21(2):63–68. doi: 10.4274/Mirt.344. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Tanzler ED, Amdur RJ, Morris CG, Werning JW, Mendenhall WM. Challenging the need for random directed biopsies of the nasopharynx pyriform sinus and contralateral tonsil in the workup of unknown primary squamous cell carcinoma of the head and neck. Head Neck. 2016;38(4):578–581. doi: 10.1002/hed.23931. [DOI] [PubMed] [Google Scholar]
- 18.McQuone SJ, Eisele DW, Lee DJ, Westra WH, Koch WM. Occult tonsillar carcinoma in the unknown primary. Laryngoscope United States. 1998;108:1605–1610. doi: 10.1097/00005537-199811000-00004. [DOI] [PubMed] [Google Scholar]
- 19.Koch WM, Bhatti N, Williams MF, Eisele DW. Oncologic rationale for bilateral tonsillectomy in head and neck squamous cell carcinoma of unknown primary source. Otolaryngol Neck Surg. 2001;124:331–333. doi: 10.1067/mhn.2001.114309. [DOI] [PubMed] [Google Scholar]
- 20.Nagel TH, Hinni ML, Hayden RE, Lott DG. Transoral laser microsurgery for the unknown primary: role for lingual tonsillectomy. Head Neck United States. 2014;36:942–946. doi: 10.1002/hed.23372. [DOI] [PubMed] [Google Scholar]
- 21.Mistry RC, Qureshi SS, Talole SD, Deshmukh S. Cervical lymph node metastases of squamous cell carcinoma from an unknown primary: outcomes and patterns of failure. Indian J Cancer India. 2008;45:54–58. doi: 10.4103/0019-509X.41771. [DOI] [PubMed] [Google Scholar]