The incidence of cutaneous metastasis from gastric cancer has been considered to be lower than 0.7% [1]. Therefore, its clinical manifestations are unclear. The clinical manifestations of cutaneous metastasis are the first signs to suggest a possible diagnosis. Thus, it is essential to clarify the manifestations of cutaneous metastatic tumors from gastric cancer. As shown by our research, delayed diagnosis was observed for 90.4% of the patients. The average interval of this delay was 3.9 ± 4.2 months. This may be a result of unfamiliarity with the clinical manifestations. This delayed diagnosis may somewhat contribute to the poor prognosis, which was calculated as 6 months by our research. One patient in our cohort [59] had a small nodule on his scalp one year postgastrectomy. No symptoms existed. The nodule slowly grew for approximately 2 years until it was removed, with a suspected diagnosis of a tumor originating from the skin. The final pathology revealed metastasis from gastric cancer. Unfortunately, further laparotomy revealed vast metastasis throughout the abdominal cavity. He died 1 year later although palliative chemotherapy was provided. An earlier diagnosis could have helped to avoid progression to advanced disease, allowing for longer survival.
The most common manifestation was an emerged nodule located in the thoracoabdominal wall. Typically, the nodule progresses slowly without pain or other obvious symptoms. Most of the nodules presented as multiple skin lesions. Some of the nodules were singular, which increased the difficulty of differentiating them from common skin lesions. The nodules varied from soft to firm and could be smooth or rough. Erysipelas-like skin lesions were the second most common manifestations. These lesions may present as a single lesion or diffuse lesions in a vast area. The presentation may lead to confusion with an infectious disease or other forms of dermatitis. Some of the patients had both characteristics. The best way to distinguish the diseases is biopsy, which is necessary to confirm the metastasis.
The thoracoabdominal wall was the most common location of cutaneous metastasis from gastric cancer, accounting for 56.9% of the cases, followed by the head and neck, with a slightly lower incidence of 54.2%. In previous reports, cutaneous metastasis has always emerged near the original organ. However, a difference was observed for gastric cancer. Most cutaneous metastases were distant from the incision. Some of them were diffuse all over the body. The pathological result of the primary tumor was more likely to be carcinoma with poor differentiation. A total of 87.5% of the gastric cancers belonged to this subset, while 66.1% were signet ring cell carcinoma. This characteristic may be considered as a high-risk factor for cutaneous metastasis. A painless nodule or erysipelas-like skin lesion that emerges in a patient who is diagnosed with poorly differentiated gastric cancer, especially signet ring cell carcinoma, should warrant high suspicious.
The diagnosis of cutaneous metastasis was not complicated. Following the initial suspicion, a biopsy should be performed. This was necessary for the final diagnosis. Immunohistochemistry was indispensable. Two metastatic tumors presented inconsistent differentiation statuses from the primary tumor. One case presented as a moderately differentiated tumor, while the metastasis presented as signet ring cell carcinoma. Another patient presented with signet ring cell carcinoma and metastasis without signet ring cell carcinoma. Nonetheless, their molecular characteristics were the same as those of the original tumor. Therefore, the diagnosis of metastatic tumors was based on immunohistochemistry results.
As later stages, treatment of cutaneous metastasis from gastric cancer is troublesome. Systemic chemotherapy was the most common strategy. The chemotherapeutic pattern varied according to NCCN guidelines. Fluorouracil and platinum were the most commonly used regimens. Forty-two patients received chemotherapy, and very few had an objective response. Most of the included patients had multiple skin lesions or extracutaneous metastases. Due to heterogeneity among lesions, the metastases in different locations had varying sensitivity to chemotherapy. The presence of multiple metastases increases the difficulty of treating the disease. Resection of the cutaneous metastases was considered to be effective in certain patients who had local skin lesions. Eight patients underwent radical resection of the cutaneous metastases with significant longer survival. All of them had local metastasis without extracutaneous disease. Local treatment to the affected skin presented promising results in alleviating symptoms. Intratumour injections of IL-2 could rapidly shrink the metastatic tumor with symptomatic alleviation. One patient received an injection of IL-2 [36]. His cutaneous metastasis shrank in 10 days. This injection had no effect on the extracutaneous metastases. The patient died 2 months later due to systemic disease deterioration.
The prognosis of gastric cancer with cutaneous metastasis is poor. The calculated median overall survival was 6.0 months. The poor prognosis is partially due to the delayed diagnosis, partly because of the unfamiliarity with such metastases. The insensitivity to chemotherapy contributed greatly to the poor prognosis. Radical resection of cutaneous metastasis may prolong survival. In total, 8 patients underwent radical surgery, for whom 5 had detailed description of their survival outcomes. The other 3 patients who underwent radical surgery were only described as having a long survival period. The median overall survival of those 5 patients was 48.0 months. Significantly longer survival was observed for those who underwent radical surgery. The longest survival in our cohort was 55 months, followed by 48 months and 40 months. All of those patients had local skin lesions without extracutaneous tumors. It seems that patients with local cutaneous metastasis and no extracutaneous metastasis may benefit from surgery.
The limitation of this paper is obvious. All the data were collected from published cases. Bias exists among the reports. Few reports had adequate information. Several reports were focused on pathological diagnosis or imaging characteristics without oncological data. Regarding survival, few studies had a clear follow-up strategy, and most reported only survival time. The follow-up was not adequate. All of these factors decrease the accuracy of the analysis. We hope that future cases will be reported with complete information.