The objectives of this study were to evaluate survival in 141 patients with stage II–IV oral squamous cell carcinoma (OSCC) treated with preoperative intra-arterial chemotherapy with docetaxel, cisplatin, and peplomycin combined with intravenous chemotherapy using 5-fluorouracil (IADCPIVF) via the superficial temporal artery, and to clarify the prognostic factors. The study population included 59 patients with stage II OSCC, 34 with stage III, and 48 with stage IV. After IADCPIVF, 139 patients underwent surgery; minimally invasive surgeries (MIS) including excisional biopsy were performed on 96 patients with a remarkably good response to IADCPIVF. The primary tumour response rate was 99.3% (complete response rate 56.7%, good partial response rate 17.0%, fair partial response rate 25.5%). Additionally, there were no serious adverse events associated with IADCPIVF. The 5-year overall survival rate was 74.6% (stage II 83.6%, stage III 72.7%, stage IV 64.8%). In the multivariate analysis of survival, T classification and clinical tumour response were significant prognostic factors. Eight (8.3%) of the patients who received MIS had primary recurrence and six were salvaged. In conclusion, IADCPIVF is safe and efficacious for treating OSCC, and MIS could reduce the extent of primary tumour resection in the case of a remarkably good response.
Intra-arterial chemotherapy for head and neck cancer treatment with radiation and intra-arterial cisplatin (RADPLAT) was first reported by Robbins et al. in 1996 . This is a combination of radiation therapy and concurrent intra-arterial chemotherapy using the Seldinger technique to administer high-dose cisplatin (CDDP) via the femoral artery. Since then, favourable treatment outcomes have been demonstrated . However, while this treatment method has an excellent anti-tumour effect, there have been reports of serious complications, such as cerebrovascular accidents and cranial nerve disorders. Its application is primarily limited to patients with locally advanced cancers, and elderly patients as well as those in generally poor physical health are excluded .
An approach in which a retrograde catheter is placed in the superficial temporal artery (STA) was first reported in 1959 . Additionally, there have been recent reports of superselective placement of the catheter directly into the target artery via the STA, which have shown excellent treatment outcomes . A long-term indwelling catheter enables the administration of a single or multiple anticancer agents continuously or frequently, which is an advantage not found when using the Seldinger technique via the femoral artery. Furthermore, there is a lower risk of cerebrovascular accident, allowing the retrograde catheter approach to be applied in elderly patients.
The following treatment strategy involving intra-arterial chemotherapy using the retrograde catheter approach was designed to establish a highly safe and effective treatment for oral cancer. (1) The treatment regimen included multiple anticancer drugs that are commonly used for the treatment of head and neck cancers. Docetaxel (DOC), CDDP, and peplomycin (PEP) were administered continuously via the artery, and 5-fluorouracil (5-FU) was administered systemically at low doses over long periods of time. The usefulness of this intra-arterial chemotherapy, which is henceforth referred to as IADCPIVF (intra-arterial chemotherapy with DOC, CDDP, and PEP combined with intravenous chemotherapy using 5-FU), was shown in a pilot study in 2006 . (2) Since then, IADCPIVF has been performed mainly as preoperative chemotherapy to treat not only locally advanced oral squamous cell carcinoma (OSCC), but also stage II lesions, because of the low incidence of serious complications associated with this therapy. For cases in which IADCPIVF was remarkably effective, an excisional biopsy or intraoral resection of the primary lesion was then performed, without pull-through resection or reconstructive surgery whenever possible, especially in elderly patients .
In this study, the therapeutic results of 141 patients with OSCC were analysed retrospectively to investigate the prognostic factors and to evaluate the safety and efficacy of IADCPIVF as preoperative chemotherapy.
Materials and methods
This study was a retrospective analysis of 141 untreated patients with OSCC who underwent IADCPIVF as preoperative chemotherapy between March 2003 and March 2015. Eligible patients had stage T2–T4, N0–N3, and M0 lesions. T2 lesions were included if they were larger than 3 cm. The patients had no severe disorders of the vital organs such as bone marrow, liver, or kidney. Patients with a performance status of 4 (Eastern Cooperative Oncology Group, ECOG) were excluded.
IADCPIVF and the study design were approved by the Ethics Committee of Fukushima Medical University, School of Medicine (number 2481), and this two-institution study was performed using the same protocol at Fukushima Medical University Hospital and Ohara General Hospital. Written informed consent was obtained from the patients.
Method for catheter placement
Branches of the external carotid artery were examined preoperatively using three-dimensional computed tomography angiography (3D-CTA). As described in the previous report by the present investigators’ group, an intra-arterial catheter (Anthron P-U catheter, tapering type, distal portion 2.7 Fr; Toray Medical Co., Ltd, Tokyo, Japan) was inserted and placed selectively under local anaesthesia via the STA or occipital artery . During catheter placement in the external carotid artery, the tip of the catheter was placed beneath the bifurcation so that the majority of the contrast agent could flow into the target artery, and the oral cancer and the surrounding mucosa could be macroscopically stained with disodium indigotin-disulfonate (Indigo Carmine; Daiichi Sankyo Co., Ltd, Tokyo, Japan). During the superselective placement, the tip of the catheter was placed directly into the target artery, which enabled all agents to flow into it .
In the previous pilot study , the recommended respective total doses of DOC, CDDP, and PEP were 40 mg, 40 mg, and 10 mg/body for patients who had catheter placement in the external carotid artery, and 30 mg, 30 mg, and 10 mg/body for those who had superselective catheter placement. The total dose of 5-FU administered intravenously was 2500 mg/body for both catheter placements. The above total doses of each anticancer drug were administered according to two different time schemes, as described below.
The following 10-day regimen was used for patients who underwent IADCPIVF between 2003 and 2011 ( Fig. 1 A): via the artery, 15 or 20 mg DOC was administered continuously from day 1 to day 5, and 10 mg PEP was administered continuously from day 6 to day 10. In addition, 15 or 20 mg DOC was administered for 3 hours via the artery on day 3, and 15 or 20 mg CDDP on days 4 and 5. Also, 250 mg/day of 5-FU was administered intravenously continuously from day 1 to day 10.
In 2012, the treatment period was shortened to 5 days with no changes in the total dose, in order to lessen the patient’s burden, DOC (30 or 40 mg) was administered continuously from day 1 to day 3, and PEP (10 mg) from day 3 to day 5. In addition, CDDP (15 or 20 mg) was administered on days 2 and 3, and 5-FU (500 mg per day) was given intravenously continuously from day 1 to day 5 ( Fig. 1 B). A second course of treatment was administered 4 weeks after the first course when necessary.
As concurrent radiation therapy, a dose of 30–40 Gray (Gy) was administered early after the introduction of IADCPIVF, even in T2 cases. However, due to the possibility of severe mucositis or bleeding during surgery, the application of concurrent radiation therapy was limited to cases in which surgery was not possible or rejected, and to T3 and T4 cases in which invasion into the bone or muscular layer was prominent.
The response evaluation criteria given in the guidelines for solid tumours of the European Organization for Research and Treatment of Cancer, National Cancer Institute of the United States, and National Cancer Institute of Canada were used to determine the treatment effects . The largest diameter of the tumour was measured before and after IADCPIVF to determine the response to treatment. The tumour response was classified as ‘complete response’ (CR), which was complete tumour regression; ‘good partial response’ (gPR), which was a residual tumour of <10 mm; ‘fair partial response’ (fPR), which was a ≥30% reduction and a residual tumour of ≥10 mm; or ‘stable disease’ (SD), which was a <30% reduction or <20% increase in the tumour.
In addition, the Common Terminology Criteria for Adverse Events version 3.0 were used to grade adverse events .
Before IADCPIVF, marker points were placed approximately 5 mm from the indurated tumour border ( Fig. 2 A, B). Resection of the primary lesion was performed depending on the effect of IADCPIVF, using the following methods, as reported previously : (1) if the effect was CR, an excisional biopsy was performed. The excision margin was inside the marker points and included the muscle layer ( Fig. 2 C, D). (2) If the effect was gPR, the excision margin was directly on the marker points. For gingival carcinoma, the excision margin was the same as above, and a partial maxillectomy or marginal mandibulectomy was performed. These two methods were considered to be minimally invasive surgeries (MIS). In fPR and SD patients, the resection was performed >5 mm away from the marker points. In all cases, with the exception of gingival carcinoma, the presence or absence of cancer in the muscle layer and mucous membrane margin was confirmed by intraoperative histopathological examination and, if positive, additional resection was performed.
The Kaplan–Meier method was used to calculate overall survival (OS) starting from the day the cancer was treated until the patient’s death (due to any cause), and the log-rank test was used to compare survival rates between different patients.
The Cox regression model was used to perform a multivariate analysis of factors affecting OS, including age, cancer site, T classification, N classification, stage, concurrent radiation, catheter placement, chemotherapy regimen, clinical tumour response, and MIS. Differences were considered significant when the P -value was <0.05.
This study enrolled 141 patients: 85 male and 56 female, with an age range of 27–93 years (median 70 years). Twenty-seven patients were aged ≥80 years. The sites of oral cancer are shown in Table 1 . By T and N classification (Union for International Cancer Control (UICC) 2005), 76 patients were diagnosed with T2 disease, 31 with T3, 28 with T4a, and six with T4b; 91 patients were diagnosed with N0 disease and 50 with N1–N3 disease. By stage classification, 59 patients (42%) were diagnosed with stage II, 34 (24%) with stage III, 41 (29%) with stage IVA, and seven (5%) with stage IVB disease.
|Characteristics||Number of patients|
|Primary tumour site|
|Floor of the mouth||15|
The observation periods for these patients ranged between 2 and 165 months, with a median of 62 months. One patient who was followed-up for 2 months died of a disease other than OSCC. Four patients were followed-up for less than 24 months: two were unable to visit the hospital because of another medical condition and the other two were transferred to another hospital after more than 12 months of follow-up.
A retrograde intra-arterial catheter was placed in the target position via the STA or occipital artery in all 141 patients. Bilateral IADCPIVF was performed on 17 patients. One course of IADCPIVF was performed in 83 patients and two or more courses in 58 patients. The 10-day regimen was performed for 111 patients and the 5-day regimen for 30 patients.
Concurrent radiation therapy was performed in 39 (27.7%) of the 141 patients, including seven patients with T2 (9.2%), 12 with T3 (38.7%), 14 with T4a (50%), and six with T4b (100%) disease.
After IADCPIVF, 139 patients underwent surgery. Overall, 96 (69.1%) patients underwent MIS, 18 (12.9%) underwent reconstructive surgery, and 25 (18.0%) underwent surgery other than MIS or reconstructive surgery. Two patients refused to undergo surgery. The T classifications of patients who underwent MIS showed that 61 (80.3%) cases were T2, 22 (71.0%) were T3, 10 (35.7%) were T4a, and three (50.0%) were T4b. The T classifications of patients who underwent reconstructive surgery showed that four (5.3%) were T2, one (3.2%) was T3, 12 (42.9%) were T4a, and one (16.7%) was T4b.
Among 50 patients with cervical lymph node metastasis, 41 underwent neck dissection. Twenty-eight of these patients had pathologically positive nodes, while the others had negative nodes. The lesions in seven cases were evaluated as CR after IADCPIVF; therefore, they were only under observation. The remaining two patients refused to undergo surgery.
Treatment efficacy was evaluated for all 141 patients. The primary tumour response rate was 99.3%; the CR rate was 56.7%, gPR rate was 17.0%, fPR rate was 25.5%, and SD rate was 0.7%. Pathological CR was found in 64 (46.0%) of the 139 patients ( Table 2 ).
|Number of patients|
n = 52
|Floor of the mouth
n = 15
n = 26
n = 23
n = 25
n = 141