Discussion
In this study, the incidence of lung metastases among patients with newly diagnosed hypo-pharyngeal cancer was calculated. We found that there was no significant difference in the incidence of lung metastases in the primary site of the tumor. Unsurprisingly, females were less likely to have lung metastases. Since many of the female patients’ lung metastases status at diagnosis was unknown, not applicable or did not record follow-up times and these patients were excluded, the true incidence of lung metastases in female patients with hypo-pharyngeal cancer is likely underestimated by the results of this study. In addition, we are pleased to report for the first time that for each 10 mm increase in tumor size, the odds of having lung metastases increases by 6.6%. This result may allow us to further predict the probability of lung metastases based on tumor size. High histological grade III/IV had the highest incidence of lung metastases in this study and was consistent with the common knowledge. Moreover, we discovered that patients with lung metastases were more likely to have other metastases. This suggests that when we find lung metastases, we should perform further tests such as: MRI, B-ultrasound and bone scan to exclude the possibility of other metastases. The development of PET-CT makes finding other metastases sites such as the aforementioned ones, much simpler.
Since hypo-pharyngeal cancer has a relatively lower incidence rate, research is scant on the topic of lung metastases of this cancer. Even research on head and neck squamous cell cancer is rare. Ampil, et al3. reported 37 patients with treated head and neck cancer had lung metastases, but did not mention the incidence rate. Marc Mareel, et al 4reported incidences of lung metastases in head and neck cancer patients at 10.8% of all the studied patients and 78% of all distant metastasis patients. They also found that the prevalence of distant metastases in hypo-pharyngeal cancer was 20.5%. This result was somewhat different from our study. The incidence of lung metastases in the Mareel study was only 4.7% of all the studied patients and the incidence of all distant metastases was 6.3% in this study. This discrepancy may be because the Mareel study only sampled from a single center’s study and did not focus on hypo-pharyngeal cancer. Our study using SEER data includes approximately 30% of the United States population, the incidence proportions we describe are highly generalizable and more likely to be reflective of the population compared to previously published data focused primarily on patients treated at academic cancer centers. Macherey, et al5. performed a meta-analysis/review of all papers published between 2000 and 2014; they found lung metastases develop in 1.9-13% of head and neck cancer patients. The results of our study seem agree with Macherey’s findings.
The study conducted by Marc Mareel, et al also showed 47% of patients died from their primary tumors during follow-up of which 52% of patients had distant metastases, however the tumor site and distant metastases site were not separated in Mareel’s study. In our study only 28.1% of patients died as a result of their first tumor; of those individuals, 57.8% of patients had lung metastases. Compared to Mareel’s study, our study focused on hypo-pharyngeal cancer and lung metastases; it can truly illustrate the effect lung metastases of hypo-pharyngeal cancer has on survival. P. Pracy, et al. reported the overall 5-year disease specific survival rate in hypo-pharyngeal cancer patients as approximately 30–35% in the United Kingdom. By reviewed articles of head and neck cancer, Macherey, et al. reports a 58% 5-year survival rate 5. In our study, 71.9% of patients were alive or succumbed to other causes instead of the first tumor after the 5-year follow-up. The reported 2-years over survival(OS) in head and neck cancer with distant metastases was 15% in Belgium, while in our study, survival was close to 40% in hypo-pharyngeal cancer patients with lung metastases and the survival probability gradually decreased to 20% after the 5-year follow-up. Our results were slightly different from the studies in the United Kingdom and Belgium, because the population composition may be different in three countries. A study including a broader range of countries could account for this discrepancy4,6.
As is shown in previous studies, a lot of risk factors will affect survival time in head and neck cancer7. Among the entire cohort, although the tumor primary site in the pyriform sinus didn’t have significantly lower odds of lung metastases, they indicate a lower risk associated with all-caused mortality (HR:0.83, 95% CI:0.69-1.00, P =0.048). This suggests that the primary site of tumor is still worthy of further research. African Americans not only have significantly greater odds of lung metastases, but also suggests a higher risk associated with all-caused mortality(HR:1.89, 95% CI:1.49-2.40, P <0.001). This finding presents a bad prognosis for African Americans who are diagnosed with hypo-pharyngeal cancer. Tumor size is an undoubted risk factor for mortality, for each 10 mm tumor size increase, the risk of death increases by 3.9%. It suggests that a closer follow-up study is needed for patients with larger tumors.
Lung metastases is often caused by cancers of peritoneal organs, including gastric cancer, colon cancer, pancreatic/biliary cancer, ovarian cancer, and uterine cancer8. Most of these cases are adeno-carcinomas; only a few cases are derived from squamous cell cancer9-12. In contrast to these studies, we found more patients having lung metastases with newly diagnosed hypo-pharyngeal cancer. We revealed that race, histological grade and tumor size were associated with having lung metastases at diagnosis. In the future we should be wary of these factors.