Mucosal melanoma of the head and neck

Introduction

Head and neck mucosal melanoma (HNMM) is a rare malignant tumor worldwide, but is relatively common in Japan, comprising 6% of all melanomas (1). Resection with a clear margin is emphasized as the most important factor for good local control and better prognosis. However, radical surgery is not possible in certain cases as many important organs may be located near the tumor. In addition, effective chemotherapy has not yet been established, and a common therapeutic strategy has not been clarified excluding curative surgical resection.

This study was performed to analyze 13 cases of MMHN treated at the Yokohama City University School of Medicine from June 1992 to November 2010.

Patients and methods

The medical records of 13 MMHN patients treated at the Yokohama City University School of Medicine from June 1992 to November 2010 were retrospectively reviewed. Collected data included gender, age, primary site, extent of the lesion, nodal involvement at presentation, metastasis at presentation, treatment contents, positive or negative surgical margins and treatment outcome.

Patients consisted of 3 males and 10 females. Median age was 60.8 years (range 39–78). Common symptoms were epistaxes (38.5%) and nasal obstruction (30.8%). In terms of the primary site, tumors in 5 cases were in the paranasal cavities (maxillary sinus in 2 cases, ethmoid sinus in 2 cases, sphenoid sinus in 1 case) and 8 cases in the nasal cavity. Staging according to the Ballantyne's staging system was applied (1). Stage I: 11 cases showed lesional type tumors. Stage II: 1 case had cervical nodal metastasis only. Stage III: 1 case had distant metastases. Furthermore, Stage I was classified into two types. Tumors in 5 cases in Stage Ia were limited to one anatomical lesion, and those in 6 cases in Stage Ib extended to the adjacent lesion.

The median follow-up period was 48 months (range 10–115). Survival rates were expressed using the Kaplan-Meyer method and the Wilcoxon test was used to evaluate the difference between each group (Table I).

Table I. Characteristics of 13 patients with mucosal melanoma of the head and neck.

Table I.

Characteristics of 13 patients with mucosal melanoma of the head and neck.

Characteristics	No.
Age (years)	39–78 (average 61)
Gender
Male	3
Female	10
Sites
Nasal cavity	8
Sinonasal cavity	5
Maxillary sinus	2
Ethmoidal sinus	2
Sphenoidal sinus	1
Tumor stage
Stage Ia	5
Stage Ib	6
Stage II	1
Stage III	1

Surgical resection

Ten out of 13 patients underwent definitive surgery as initial treatment. Three patients did not undergo surgical operation due to 1 case each of liver metastases, orbital and skull base invasion, and skull base invasion.

Chemotherapy and radiotherapy

In addition to surgery, chemotherapy and/or immunotherapy were carried out in the initial treatment course (Table II). Chemotherapy was based on the so-called DAV therapy: Dimethyltriazeno-imidazole carboxamide (DTIC; 80–140 mg/m2), amino methyl pyrimidinyl methyl chlorethyl nitrorosourea hydrochloride (ACNU; 50–80 mg/m2) and vincristine (VCR; 0.5–0.8 mg/m2). Cisplatin with DAV was administered for 3 patients and interferon-α with DAV was applied in 5 cases (Table III).

Table II. Characteristics and outcomes of each patient.

Table II.

Characteristics and outcomes of each patient.

Pt. no.	Age	Gender	Major complaint	Site	Stage	Primary treatment	Surgical margin	Adjunct LAK cell therapy (no. of times)	Outcome (follow-up period, months)
1	45	F	Cheek pain	MS	Ib	C→O→R	−	−	Death (13)
2	67	M	Epistaxis	N	Ia	C→O→I	−	−	Death (38)
3	54	M	Epistaxis	ES	Ia	C→I		−	Death (35)
4	74	F	Nasal obstruction	N	Ia	O→I	+	+ (32)	Death (64)
5	54	F	Nasal obstruction	N	Ia	O→I	−	+ (6)	Death (42)
6	68	F	Nasal obstruction	N	Ia	C→O→I	−	+ (13)	Death (115)
7	40	F	Epistaxis	N	Ib	C→O→I	+	+ (15)	Death (35)
8	56	M	Epistaxis	N	Ib	O→I	+	+ (23)	Death (89)
9	74	F	Nasal obstruction	N	Ib	O→C→I	−	−	Alive (65)
10	40	F	Asymptomatic	SS	III	R→C→I		+ (22)	Alive (62)
11	72	F	Exophthalmos	MS	Ib	C→I		−	Alive (31)
12	69	F	Neck tumor	MS	II	C→O→I	−	−	Alive (31)
13	78	F	Epistaxis	N	Ib	C→O→I	−	+ (6)	Alive (10)

[i] MS, maxillary sinus; ES, ethmoidal sinus; SS, sphenoidal sinus; N, nasal cavity. C, chemotherapy; O, operation; I, immunotherapy.

Table III. Chemotherapy regimens.

Table III.

Chemotherapy regimens.

DAV + CDDP	3
DAV	2
DAV-feron	5

Radiotherapy during initial treatment was carried out in 2 cases. Postoperative radiotherapy (50 Gy) was carried out for 1 case and heavy particulate radiotherapy was carried out for 1 inoperative case.

Immunotherapy

As for immunotherapy, lymphokine-activated killer (LAK) cell transfer therapy (2,3) was carried out for 7 patients as adjunctive treatment after initial treatment. LAK cell therapy populates the self-activated lymphocytes intravenously with interleukin-2 (IL-2). Grimm et al (4) and Rosenberg et al (5) demonstrated an increasing trend in the survival rate when IL-2 was administered with LAK cells in patients with advanced cancer. Blood (40–50 ml) was collected from the patients. Approximately 109 lymphocytes were prepared after culture with the anti-CD3 antibody and IL-2, and subsequently CD-8-positive killer (LAK) cells were induced and proliferated. The average number of adjunctive LAK cell treatments was 16 (range 6–32). Many patients received the therapy once a week or once every 2 weeks. The number of cells injected each time was 5×108 to 5×109.

Results

The overall 5-year, cause-specific survival rate was 56% (Fig. 1). Each rate showed a better outcome than those of other reports (6,7). As for the operative therapy, the 5-year disease-free survival rate in margin-positive surgery (n=3) was 0%, whereas it was 50% in margin-negative surgery (n=7) (p=0.21) (Fig. 2). In 7 cases receiving adjunctive LAK cell therapy, the 5-year cause-specific survival rate was 66%, while that in 6 cases without adjunctive LAK therapy was 33% (p=0.43) (Fig. 3). Although a statistical significance was not recognized, LAK therapy is suggested to improve prognosis of MMHN.

Figure 1. The 5-year cause-specific survival rate was 56%, which showed a better outcome than those of other reports.

Figure 2. Disease-free survival rate of patients with positive/negative surgical margins. The 5-year disease-free survival rate in margin-positive surgery was 0%, whereas it was 50% in margin-negative surgery, indicating that curative surgery contributes to better disease control.

Figure 3. Cause-specific survival rate in patients with/without adjunctive LAK cell therapy. In cases receiving adjunctive LAK therapy, the 5-year cause-specific survival rate was 67%, while in cases without adjunctive LAK therapy, it was 33%.

Discussion

The incidence of malignant melanoma in Japan is low compared to Western countries. However, regarding the incidence of MMHN, in Western countries it accounts for only 1.7% of all cases of melanoma, whereas it accounts for 23.3% in Japan (8,10). MMHN arises from aberrant melanocytes during fetal life. On the other hand, a difference in sensitivity to ultraviolet rays is assumed to cause skin melanoma. Therefore, racial difference is considered to be associated with a high incidence of melanoma of the skin in Western countries.

With regard to gender and MMHN, certain reports have shown a male-to-female ratio of 1:1 to 2:1 (8,10). In this study, the ratio was 3:10. Regarding the primary site of MMHN, various studies have reported a rate of occurrence of 50% in the oral cavity and 35% in the sinonasal cavity (10), or a high occurrence in the sinonasal cavity (11). In this study, 8 cases (61%) originated from the nasal cavity and 5 cases (39%) from the paranasal cavities.

A treatment modality for MMHN has yet to be established. For resectable lesions, complete removal with surgical margins, such as a basilar operation, is generally the most optimal course. Regarding chemotherapy for MMHN, a DTIC-based combination regimen is often applied (12). Penel et al reported that positive surgical margins were a risk factor in univariate analysis, and they suggested that clear margins appeared to predict a more favorable outcome (13). Other reports also revealed that the quality of the margins was linked to a better overall survival rate (10,14). Thus, operative therapy is the first choice for curative treatment. In our study, the 5-year disease-free survival rate in margin-positive surgery (n=3) was 0%, whereas it was 50% in margin-negative surgery (n=7) (p=0.21) (Fig. 2).

On the other hand, recent reports have shown almost the same effectiveness in terms of local control by high-dose fractionated radiotherapy in comparison to curative surgery (7,15). Wada et al reported that high-dose per fractionated radiotherapy doses (≥3 Gy) were associated with better prognosis for both local control and survival (9). The effect of postoperative radiotherapy has also been the focus of research (14). Temam et al demonstrated that patients with early T-classification tumors who received postoperative radiotherapy had a better local disease-free survival compared to patients with late T-classification tumors who did not receive postoperative radiotherapy (16). Thus, the role of radiotherapy, such as high-dose fractionated radiotherapy and carbon ion radiotherapy, is expected to be considerably more important for functionally and cosmetically operable cases (9,17,18). However, radiotherapy treats the local tumor, but does not inhibit distant metastasis (14,16). A combination treatment with systemic therapies, such as chemotherapy and immunotherapy, is indispensable for the improvement of the outcome of MMHN (14,16).

As for immunotherapy, LAK cells were administered intravenously with IL-2 to retain its activity in attacking cancer cells. Grimm et al (4) and Rosenberg et al (5) found an increased trend in the survival rate when IL-2 was administered with LAK cells in patients with melanoma. Seven patients received LAK cell therapy as adjunctive therapy after initial treatment. The 5-year cause-specific survival rate of this group was 67%, while that of 6 cases without LAK cell therapy was 33% (p=0.43). Due to the small population, a statistical significance could not be obtained. However, based on the results, LAK cell therapy is suggested to improve prognosis of MMHN (Fig. 3).

Because of the small size of most reported series and their retrospective nature, the effect of various treatment strategies for MMHN has been difficult to evaluate.

Surgical treatment has been the primary treatment modality. However, as wide surgical resection in the head and neck region is often difficult, radiotherapy has been recently reported as a beneficial management modality. The high rate of distant metastasis also suggests that a systemic treatment is required. Currently, there is no role for adjuvant systemic therapy for patients who have been successfully resected. Further evaluation of the role of chemotherapy and immunotherapy is required to decrease the rates of distant metastasis and improve survival (19).

There are few reports on immunotherapy for MMHN. Based on dermatological research, malignant melanoma is targeted using immunotherapies, such as cytokines, cancer vaccines, adoptive cell transfer and cancer gene therapies, and these immunotherapies have been actively investigated (20–22). For example, Atkins et al reported that 12 responding patients out of 270 patients with metastatic melanoma treated with high-dose recombinant IL-2 therapy remained continually disease-free or progression-free from >70 to >150 months following initiation of therapy (20).

In the present study, LAK cell therapy as initial or adjunctive therapy for patients with MMHN was applied. A tendency for an improved survival rate of patients with MMHN by LAK cell therapy was recognized. Further investigation using an accumulation of cases is warranted.

Abbreviations:

References