Superselective arterial infusion chemotherapy for squamous cell carcinomas of the oral cavity: histopathologic effects on metastatic neck lymph nodes

Abstract

We have performed superselective intra-arterial infusion chemotherapy (SIC) on carcinomas of the oral cavity according to a protocol in which the distribution of the drug was evaluated by the use of a combined CT and angiography system, and the chemotherapy was combined with medium-dose conformal radiation therapy (CRT). The purpose of this study was to evaluate the pathological effect of this treatment on the metastatic neck lymph nodes (LNs). Twenty consecutive patients who had metastatic neck LNs from squamous cell carcinomas of the mouth and who underwent both SIC and CRT were included in this study, in which a total of 22 LNs were evaluated. A microcatheter was placed in the appropriate feeding artery of the tumor, such as the internal maxillary artery, facial artery, lingual artery and external carotid artery (ECA), and cisplatin (50 mg/body) was infused twice through a microcatheter. The CRT was administered with a dual-energy (4 and 10 MV) linear accelerator. The total and daily doses delivered were 30 and 2.0 Gy, respectively. Intra-arterial infusion to the LNs was divided into two groups: superselective infusion (mainly to the submandibular LNs via the facial artery, n = 10) and nonsuperselective infusion via the ECA (n = 12). The distribution of cisplatin into the LNs was confirmed by slow-infusion CT. Histopathologic effects on the LNs were evaluated on the specimens obtained during the operation and classified into five grades (0: no or minimal response: I: disappearance of less than three quarters of the tumor cells: II: disappearance of more than three quarters of the tumor cells: III: disappearance of viable tumor cells with a small amount of residual nonviable tumor cells: IV: complete disappearance of all viable and nonviable tumor cells). Grade 0 or 1 was defined as poor response and Grade II or more as good response. Twenty-three LNs from nine patients without CRT and SIC were served as control. In the superselective infusion group, all 10 LNs showed good response (response rate, 100%: grade II = 4, grade III = 3, grade IV = 3). In the non-superselective group, however, 6 of 12 LNs showed poor response (response rate, 50%: grade 0 = 2, grade I = 4, grade II = 2, grade III = 2, grade IV = 2). All 23 control LNs with no treatment showed grade 0 response. Superselective infusion seems necessary to obtain good histopathologic effects on the metastatic LNs. SIC combined with CRT can be applied to the metastatic LNs.