The nasal cavity and sinus involvement by dissemination of hepatic malignancy is rare. In the literature, renal cell carcinoma(RCC) is the most common metastatic malignance of the sinonasal tract(7). So far studies evaluating metastatic hepatocellular carcinoma in the nose have been limited to case reports and small cohorts(8–11).However, accurate diagnosis in these cases is necessary, on count of the treatment of primary carcinoma of nasal-sinus differs from the cure of metastasis.
The symptoms of patients with nasal metastatic hepatocellular carcinoma were not specific, similar to that of primary sinonasal tumors. In our study, the most frequent symptom was epistaxis. It is consistent with the fact that HCC is usually vascular(12). And compared to tumors in paranasal sinus, nasal masses were more likely to cause bleeding. Numbness and pain in face and nose were common symptoms, it could be the result of the pressure on the nerve from tumor.
The histological diagnosis of primary and secondary sinonasal carcinomas can be challenging. Metastatic HCC to the sinonasal cavity must be distinguished from other tumors such as poorly differentiated squamous cell carcinoma, the non-intestinal type sinonasal adenocarcinoma, neuroendocrine neoplasm, acinic cell carcinoma and metastatic tumors such as lung adenocarcinoma, pancreatic islet cell tumor, and poorly differentiated adenocarcinoma from esophagus, stomach, and pancreas. Morphologically, sinonasal metastases tend to spare the submucosa, instead preferring to develop in the mucosa. Indispensably, immunohistochemical staining is required to establish the eventual diagnosis. Use of stains is especially vital in differentiating between primary sinonasal malignancies and metastatic carcinoma to the nasal cavity and sinus, as there exists significant histopathologic overlap between these entities. P63 and P40 stains positively in almost primary sinonasal squamous cell carcinoma which is the most common form in nasal cavity and paranasal tissues(13). The immune phenotype of the non-intestinal type sinonasal adenocarcinoma is characterized by positivity for cytokeratin 7(14). Positive DOG-1 staining would support primary acinic cell carcinoma(15). Neuroendocrine neoplasm always showed phenotypical expression of neuroendocrine markers such as CD56, synaptophysin and chromogranin(16). In addition, cytokeratin 7 almost negative in metastatic hepatocellular cells, which is different from other adenocarcinoma cells. Above all, it is important to acquire medical history in diagnosis of non-primary tumors. Necessarily, the detection of four biomarkers, hepPar-1, AFP, GPC3 and Arg-1, plays an vital part in diagnosis of metastatic hepatocellular carcinoma. Among these, HepPar-1 was the most sensitive marker with the positive rate as 75.00%(9/12). And GPC-3 was positive in 50%(6/12) cases. However, AFP as well as Arg-1 was not sensitive enough and the positive rate in tumors was 8.33%(1/12). tumor cells expressed at least one of the four indexes. HepPar-1 was proved to be the most helpful biomarker, whereas the AFP expression is not helpful in the tumor cells. In our cohorts, either HepPar-1 or GPC-3 was expressed in all 12 cases. It is effective to applicate these two immunohistochemical indexes synthetically. The detection of immunohistochemical strains and history taking would be required to make eventual diagnosis.
Hepatocellular carcinoma is a leading cause of cancer-related morbidity and mortality and it accounts for the second most cancer deaths in the world(17). Men are more likely to develop HCC worldwide than woman (18). In the Literature, the majority of reported cases of HCC metastatic to the sinonasal area have been in men (11). Likewise, the majority patients in our study were males. There are multiple therapeutic methods for HCC, such as surgical resections or liver transplantation, ablative electrochemical therapies, chemoembolization, radioembolization and so on(19). Even so, prognosis of HCC is drastically different with multiple determinants(20). Surgical resection is a potentially curative option for patients at an early stage, with three and five-year survival rates of 72% and 50% respectively(21). In this study, the five-year survival rate was 75.00%(3/4) in these four follow-up patients. And five patients with related follow-up information all survive more than 3 years. However, patients with sinonasal metastasis had a dismal prognosis and they all died within one to nine months. Overall, the prognosis of patients with HCC with distant metastasis is poor, and the clinicopathologic characteristics and the survival of metastatic HCC varied according to various metastatic organs(22). Zhan et al. founded that lung metastasis has a worse prognosis than bone metastasis, and the average survival time of brain metastasis is shorter than bone or lung metastasis(23). Furthermore, the median survival of HCC with bone metastasis is only 1–2 months(24). On the other hand, there are different endings about virous extranasal cancers that metastasize to the sinonasal cavity. Chang’s study revealed patients who received surgical excision of nasal septal metastatic thyroid papillary carcinoma could have a good life(25). Overall survival rates are disappointing in our study may indicate that sinonasal metastasis of HCC is a factor for poor prognosis.
Recently ,the immune checkpoint inhibitors directed PD-1 (nivolumab and pembrolizumab) have provide novel treatment opportunities for patients with HCC(26). Related to the drug effect is the expression of PD-L1 on the tumor and its associated cells. And patients with tumor PD-L1\(\ge\)1% would get better treatment benefit than those PD-L1 < 1%(27). However, significant heterogeneity was founded in the tumor expression of PD-L1 across primary and secondary HCC samples(28). Juan et al. reported a case with positive membranous staining for PD-L1 in metastatic tumor cells and tumor infiltrating lymphocytes(29). In our study, we founded the positivity of PD-L1 in these metastatic tumors was 8.3%(1/12). Regretfully, the PD-L1 expression of the primary HCC samples status was unknown because HCC slides were not available from other hospitals. Nevertheless these findings indicated a potential immunotherapy opportunities for advanced patients. Recently, research shows PD-L1 positive patients with HCC were more likely to suffer from aggressive clinicopathologic features than PD-L1 negative patients(30). We are not able to obtain the same results as to the sinonasal metastases.
There are several limitations in our study. It is difficult to get complete data and verification because this is a retrospective study. Also, the limited number of cases for such a rare condition, may prevent meaningful statistical analysis.