Purpose: Mucosal melanoma is rare and associated with extremely poor prognosis. However, standard adjuvant therapy for mucosal melanoma has not been established. We conducted a randomized phase II clinical trial in patients with resected mucosal melanoma to compare the efficacy and safety of high-dose IFN-α2b (HDI) and temozolomide-based chemotherapy as adjuvant therapy.
Experimental Design: Patients with mucosal melanoma in stage II/III after surgery were randomized into three groups: observation group (group A, surgery alone), HDI group (group B, treated with 15 × 106 U/m2/d IFN-α2b, followed by 9 × 106 U IFN-α2b), and temozolomide (200 mg/m2/d) plus cisplatin (75 mg/m2) group (group C). The endpoints were relapse-free survival (RFS), overall survival (OS), and toxicities.
Results: One hundred and eighty-nine patients were enrolled and finally analyzed. With a median follow-up of 26.8 months, the median RFS was 5.4, 9.4, and 20.8 months for group A, B, and C, respectively. Estimated median OS for group A, B, and C was 21.2, 40.4, and 48.7 months, respectively. Patients treated with temozolomide plus cisplatin showed significant improvements in RFS (P < 0.001) and OS (P < 0.01) than those treated with either HDI or surgery alone. Toxicities were generally mild to moderate.
Conclusion: Both temozolomide-based chemotherapy and HDI are effective and safe as adjuvant therapies for resected mucosal melanoma as compared with observation alone. However, HDI tends to be less effective than temozolomide-based chemotherapy for patients with resected mucosal melanoma in respect to RFS. The temozolomide plus cisplatin regimen might be a better choice for patients with resected mucosal melanoma. Clin Cancer Res; 19(16); 4488–98. ©2013 AACR.
Mucosal melanoma is rare and of extremely poor prognosis. However, standard adjuvant therapy for mucosal melanoma has not been established. The importance of our study is that we provide clinical data showing that high-dose IFN-α2b (HDI) and temozolomide plus cisplatin may be advised to patients with mucosal melanoma in stages II/III and after surgical removal of primary mucosal melanoma. The significance of the study is that temozolomide plus cisplatin regimen, likely better than HDI, may be a choice of adjuvant therapy for patients with mucosal melanoma in respect to relapse-free survival (RFS) and possibly overall survival (OS). Our study shows that both adjuvant regimens are safe and well tolerated for patients with resected mucosal melanoma. Even for patients harboring c-kit or BRAF mutations that are potential targets for targeted therapies, the temozolomide plus cisplatin regimen may be a better choice than HDI. Our trial is unique in that it addresses adjuvant therapy for a specific, uncommon subtype of melanoma.
As compared with cutaneous melanoma, mucosal melanomas are extremely rare in Caucasians (1.3% of all melanomas), whereas this is the second most common subtype in Asians (1–3). Mucosal melanomas are generally classified according to primary sites (4), and have a very poor prognosis and significantly worse outcomes than cutaneous melanoma (3–7). One retrospective study in our center reported that the median survival time was 3.58 years for mucosal melanoma, and the 5-year survival rate for mucosal melanoma (all stages of mucosal melanoma) was 26.8%, with significantly poorer survival to other subtypes of melanoma (3). Early diagnosis combined with appropriate surgical therapy is currently the only curative treatment of melanoma (8, 9). Strategies for early diagnosis of mucosal melanoma and effective adjuvant therapies for mucosal melanoma are urgently needed.
Systemic adjuvant treatment of melanoma may be considered when a patient is clinically free of disease following surgical excision of the primary high-risk tumor and is at high risk for recurrence, especially for patients with American Joint Committee on Cancer (AJCC) stage of IIB, IIC, and III (8, 10, 11). Many randomized trials have been conducted in stage II/III melanomas, most of which are for cutaneous melanomas and yield negative results in exception of high-dose IFN-α2b (HDI; refs. 11, 12). The randomized trials with HDI suggest that HDI can significantly improve relapse-free survival (RFS) but not overall survival (OS) of patients with high-risk melanoma (13–15). Meta-analyses also provided controversial conclusions about the effects of HDI on OS, but suggested that HDI may tend to improve OS in high-risk melanoma (12, 16). Temozolomide, the oral analog of dacarbazine (DTIC), shows potential advantages over dacarbazine, including oral administration as well as penetration through the blood–brain barrier. Cisplatin is an agent that can potentially enhance the activity of temozolomide. Clinical trials have shown that temozolomide with or without cisplatin is active in patients with untreated melanoma and can increase the progression-free survival (PFS) and the quality of life (17–19). However, the efficacy of temozolomide plus cisplatin as well as HDI as adjuvant therapy for patients with resected mucosal melanoma has not been evaluated.
Trials are seldom conducted for specified subtypes of melanoma, such as acral melanoma and mucosal melanoma. Several groups have explored the treatment of mucosal melanoma (20–23). However, most of these studies were retrospective, associated with small sample sizes, or were nonrandomized studies. Considering that adjuvant therapies have never been established for patients with mucosal melanoma, we conducted a randomized phase II study to compare the activity and safety of temozolomide plus cisplatin with that of HDI and observation in patients with resected mucosal melanoma.
Patients and Methods
One hundred and eighty-nine patients with pathologically confirmed diagnosis of mucosal melanoma that had been completely resected were included in this trial. The inclusion criteria were: age more than 18 years; Eastern Cooperative Oncology Group (ECOG) performance status of 0 or 1; pathologically confirmed diagnosis of mucosal melanoma in stages II and III; appropriately and completely resected primary tumor (once regional lymph nodes were involved, diagnosed by clinical or imaging examinations, lymphadenectomy was conducted); no prior systemic adjuvant therapy or regional radiotherapy; no evidence of distant metastatic disease evaluated by means of lymph nodes ultrasound, endoscopy, and ultrasound of anorectum and genitourinary tract, single-photon emission computed tomography (CT) of bone, and whole-body spiral CT or positron emission tomography-CT (PET-CT); normal bone marrow function; and adequate liver and renal function [including white blood cell (WBC) count > 3,000/mm3; absolute neutrophil count > 1,500/mm3; platelets > 100,000/mm3; serum creatinine less than two times of the upper limit of normal (ULN); bilirubin less than 1.5 times of ULN; aspartate aminotransferase (AST) and alanine aminotransferase (ALT) less than 2.5 times of ULN; international normalized ratio less than 1.5 times of ULN; and partial thromboplastin time less than ULN].
The exclusion criteria were: cutaneous melanoma or ocular melanoma or melanoma of unknown primary site; incomplete resection or primary tumor unable to be resected; a second cancer diagnosis; definite medical history of cirrhoses of the liver or autoimmune diseases; severe depression; and pregnant or lactating female.
The staging system for mucosal melanoma has not been well established. Taking into consideration of staging system of cutaneous melanoma (10) as well as that of gastrointestinal stromal tumor (24), and on the basis of experiences at our center, we defined the stages [tumor–node–metastasis (TNM) staging] as following: T1 for tumor invading submucosa, T2 for tumor invading muscularis propria, T3 for tumor invading adventitia, T4 for tumor invading adjacent structures, N1 for 1 regional metastatic node, N2 for 2 to 3 regional metastatic nodes, N3 for 4 or more regional metastatic nodes, M1 for distant metastases, T1–2N0M0 for stage I, T3–4N0M0 for stage II, TxN1–3M0 for stage III, and TxNxM1 for stage IV. The invasion of mucosa and lymph nodes were determined by pathologic hematoxylin and eosin (H&E) staining of melanoma specimen obtained during surgery and confirmed by CT or PET-CT scanning.
We used a simple randomization method to allocate patients into three groups by a ratio of 1:1:1 with permuted blocks of size of 3 patients: observation group (group A), the 1-year HDI group (group B), and the temozolomide plus cisplatin group (group C). Patients in group B were treated with i.v. 15 × 106 U/m2/d IFN-α2b (Intron; Merck & Co.) on days 1 to 5 each week for 4 weeks, and then followed by s.c. 9 × 106 U IFN-α2b three times per week for 48 weeks, which has been confirmed by previous trial to be tolerable and may be the optimal dose intensity for Chinese patients and Europeans (25, 26). Patients in group C were treated with per os 200 mg/m2/d temozolomide (Temodal or Temodar; Merck & Co.) on days 1 to 5 plus i.v. 75 mg/m2 cisplatin (cisplatinum; Qilu Pharmaceutical) divided into 3 days, which was repeated every 3 weeks for six cycles. The CONSORT diagram of the study was shown in Supplementary Fig. S1.
The trial was registered at Chinese Clinical Trial Register (CHiCTR Trial Registry ID: ChiCTR-TRC-11001798). Considering the small sample size and exploratory feature of this trial, and given the fact that adjuvant therapy (especially chemotherapy) for mucosal melanoma had not been fully evaluated in clinical trials, we defined this single-center trial as a phase II clinical trial to evaluate the efficacy and safety of adjuvant therapies in patients with mucosal melanoma. Results were presented in part as oral presentation at the 47th Annual Meeting of the American Society of Clinical Oncology, June 1–5, 2012, Chicago, IL. This single-centered phase II trial was approved by the Institutional Review Board and was conducted in accordance with the ethical principles originating from the Declaration of Helsinki and with Good Clinical Practice as defined by the International Conference on Harmonization. All patients (or their legal representatives) gave written informed consent before enrollment. Patient safety and treatment efficacy were monitored by an independent Data Monitoring Committee composed of clinical experts in our hospital.
Evaluation of recurrence and metastasis
Ultrasound of superficial lymph nodes, CT scans of chest, abdomen, and pelvis, and endoscopy of primary sites were conducted every 2 months in the first year, every 4 months during the second and third year, and every 6 months for the last duration of the trial. MRI scans of the head and single-photon emission CT of bone were obtained every 6 months during the initial 3 years of the trial and then repeated annually. These examinations were carried out regularly to monitor recurrence, which was further confirmed by biopsy when suspected recurrence or metastasis. Appropriate examinations including imaging scan were conducted to evaluate the presence of metastasis as clinically indicated. For objective evaluation of recurrence or metastasis, the obtained data were independently reviewed by 3 medical imaging experts.
Evaluation for adverse effects or toxicity
For patients in group A, blood tests, liver function, and blood urea nitrogen and creatinine tests were conducted every 4 weeks. For patients in group B, the above-mentioned tests were conducted before treatment, weekly during 4-week induction treatment and monthly during subcutaneous maintenance treatment. For patients in group C, the above-mentioned tests were monitored weekly following the first daily dose of each cycle to determine the dose of temozolomide and cisplatin for subsequent cycles.
For patients treated with HDI, treatment was withheld until the patient recovered in case grade 3/4 toxicities occurred, and a 33% dose reduction was conducted for subsequent treatment. When the same grade 3/4 toxicities occurred for second time, a 66% dose reduction was instigated. In case of febrile grade 3/4 neutropenia, the administration of granulocyte colony-stimulating factor (G-CSF) was permitted. Once grade 4 hepatotoxicity occurred, treatment was discontinued immediately. For patients treated with temozolomide plus cisplatin, the dose was reduced by 25% of temozolomide and cisplatin, when grade 3/4 hematologic toxicities occurred. A 50% dose reduction was required in case grade 3/4 nonhematologic toxicities occurred. If the laboratory parameters and/or clinical signs failed to return to normal or further deteriorated within 4 weeks after reduction of the dose, the treatments were interrupted.
DNA preparation and mutation screening
Genomic DNA was extracted from formalin-fixed, paraffin-embedded sections using the QIAamp DNA FFPE Tissue Kit (Qiagen). To detect hotspot mutations, we amplified exons 9, 11, 13, 17, and 18 of the c-kit gene, exons 11 and 15 of the BRAF gene, and exons 1 and 2 of NRAS gene, by PCR in at least two separate preparations of genomic DNA as described previously (27, 28). PCR products were directly sequenced by them using Big Dye Terminator sequencing chemistry on an ABI3130 automated sequencer (Applied Biosystems).
Taking previous reports into consideration and on the basis of experience at our center (29, 30), the RFS of mucosal melanoma in observation group was expected to be approximately 4 months. The aim of this trial was to detect a 100% improvement in RFS in group B as compared with that in group A (with an expected RFS of 8 months; α = 0.05; power = 0.8), and a 100% improvement in RFS in group C as compared with that in group B (with an expected RFS of 16 months; α = 0.05; power = 0.8). At each analysis, a step-down testing procedure was used. We compared the group A and B firstly at a significance level equal to 0.05. If the comparison was significant, we compared group B and C at a significance level equal to 0.05; if not significant, group B and C would not be compared. On the basis of the difference of group A and B, the calculated minimum sample size was 49 in each group. On the basis of the difference of group B and C, the calculated minimum sample size was 63 in each group. Because of the 1:1:1 design, we needed 63 patients in each group [estimated by Software (PASS 2008 Home version)]. All the patients needed to be randomly assigned followed by 2 years of observation, with an allowance of 5% dropout and a planned accrual time of 30 months. All patients had reached at least 24 months of follow-up time until the cutoff of data in July 2011.
The primary endpoint was RFS time. Secondary endpoints included the toxic effects as well as OS time. RFS time was defined as the time from the date of randomization to the documentation of any of the following failures: local or distant recurrence of disease (evaluated by ultrasound of superficial lymph nodes, CT scans of chest, abdomen, and pelvis, or lymph node biopsy), death from any cause, or completion of follow-up. OS time was defined from the date of randomization to the date of death as a result of any cause.
All the statistical analyses were conducted using SPSS 13.0 software. Categorical data are described using frequencies or percentages. Continuous data such as age are described using median (range) for normally distributed data. Adjusted χ2 test, χ2 test, or Fisher exact test were used to compare baseline data and toxicity between groups. The probability of RFS and OS were estimated by the Kaplan–Meier method, and the significance was evaluated by the log-rank test. A Cox proportional hazards model was used to obtain a point estimate of the HR for group C relative to group B, along with a 95% confidence interval (CI) for the HR for each time-to-event analysis. All statistical analyses were two-sided and significance was assigned at P < 0.05.
Patient characteristics and treatments
Between January 2007 and July 2009, a total of 189 patients (63 patients in each group) were enrolled in the trial. The median time from surgery to adjuvant treatment was 21 days (range, 16–27 days), 24 days (range, 17–30 days), and 22 days (range, 14–28 days) for group A, B, and C, respectively. No patient was lost to follow-up in group A. One patient in group B, who only received 1-month of HDI therapy, was lost to follow-up. Two patients in group B withdrew from the study because of repetitive grade 4 hematologic toxicity or grade 4 hepatotoxicity. In group C, 1 patient who received two cycles of chemotherapy was lost to follow-up, and 1 patient withdrew from the study because of repetitive grade 4 nausea and vomiting (for study flow chart; see Supplementary Fig. S1). In group B, all 63 patients had finished 4-week HDI treatment. Almost all the patients had finished six-cycle chemotherapy in group C.
All the patients were in stages II/III of mucosal melanoma, and patients in stage I were unavailable possibly due to the difficulty of diagnosis of mucosal melanoma at early stage. Although a simple randomization method was used for allocation, the distribution of patients in stage II and III was similar and balanced between the three groups. Patient characteristics were well balanced between the groups (Table 1).
Analyses were done in the intention-to-treat (ITT) population (Supplementary Fig. S1). Patients were observed for a median follow-up time of 26.8 months (range, 5.9–53.9 months); relapse of mucosal melanoma was observed in 167 patients (88.4%), with 63 events in group A (100%), 58 events in group B (92.1%), and 46 events in group C (73.0%). For those patients with recurrence at primary sites, appropriate surgery was advised, whereas for patients with metastasis, regimens appropriate for advanced melanoma were prescribed after surgery. The detailed outcomes of recurrence and metastasis for the three groups at the end of the first and second year or follow-up were listed in Table 2. The median RFS was 5.4 months (95% CI, 4.2–6.6 months) in group A, 9.4 months (95% CI, 7.9–10.9 months) in group B, and 20.8 months (95% CI, 17.9–23.7 months) in group C, which suggested that both HDI (P < 0.001) and temozolomide plus cisplatin (P < 0.001) regimens could improve RFS when compared with observation group. Figure 1A shows the RFS curves of three groups, suggesting that the temozolomide plus cisplatin regimen tends to more significantly improve the RFS of patients with resected mucosal melanoma, as compared with the HDI regimen (P < 0.001).
By July 2011, death was recorded in 98 patients (51.9%), with 47 events in group A (74.6%), 30 events in group B (47.6%), and 21 events in group C (33.3%). The estimated median OS was 21.2 months (95% CI, 15.8–26.6 months) in group A, 40.4 months (95% CI, 32.5–48.3 months) in group B, and 48.7 months (95% CI, 41.8–55.6 months) in group C, which suggested that both HDI (P < 0.001) and temozolomide plus cisplatin (P < 0.001) regimens could improve OS when compared with observation group. Figure 1B shows the OS curves for the three study groups, indicating that temozolomide plus cisplatin tends to more significantly improve OS as compared with HDI group (P = 0.009).
Target therapies using kinase inhibitors or monoclonal antibodies have shown significant improvements of clinical outcomes in metastatic melanoma (31–34). Recently, ipilimumab has been evaluated in advanced mucosal melanoma in an Italian cohort, showing that ipilimumab is active in some patients with mucosal melanoma, with a disease control rate about 23.1% (35). However, it is still largely unknown about the effectiveness of these therapies as adjuvant therapy in patients with mucosal melanoma. To understand the proportion of the patients we enrolled who might be candidates for these recently developed therapeutic approaches, we determined prevalence of these oncogenes in our population. The overall frequencies of c-kit, NRAS, and BRAF genetic mutations in this trial were 13.6% (24 of 176), 5.7% (10 of 176), and 16.5% (29 of 176), respectively, comparable with our previous reports (27, 28). Although these groups represented a small subpopulation within already small cohorts, there seem to be no differential effect of either HDI or temozolomide plus cisplatin regimens in those patients with or without DNA mutations as compared with the observation group (Table 3). Therefore, there is no ample evidence that gene mutation status could affect the efficacy of adjuvant therapy, and efficacy of targeting inhibitors as adjuvant therapy needed further trials.
Unstratified exploratory subgroup analyses of both PFS and OS for patients in groups B and C were conducted to provide a suggestive tendency for clinical prescription, although the sample sizes for several analyzed risk factors were small (Fig. 2A and B). Regardless of age, gender, primary sites, stage, and genetic mutations, temozolomide plus cisplatin could reduce the risk of recurrence (ranging from 48% to 97%) and the risk of death (ranging from 41% to 48%). Among patients with mucosal melanoma of head/neck origin, the reduction in risk of recurrence was 79%, more than that in patients with anorectum and genitourinary mucosal melanoma, suggesting that temozolomide plus cisplatin might have a higher impact for patients with head and neck mucosal melanoma. Among patients with stage II and III mucosal melanoma, temozolomide plus cisplatin led to the reduction of 64% and 90% in risk of recurrence, indicating the temozolomide plus cisplatin regimen might be effective in both stages. For patients with c-kit or BRAF mutations, temozolomide and plus cisplatin led to a reduction of 88% and 97% in risk of recurrence, respectively. For the subgroup analyses of OS, the risk of death was comparable between baseline risk factors in except of stage (59% reduction for patients in stage II vs. 75% reduction for patients in stage III). Our study may be a preliminary and exploratory study for these factors as prognostic indicators in adjuvant chemotherapy and HDI, which are definitely needed to be investigated in further trials.
The adverse events are listed in Table 4. All 189 patients had adequate toxicity data available and thus were available for the safety analyses. For patients in group A, as the observation group alone, they had little intervention-related side effect naturally. Apparently, the rates of the toxicities in group B such as fever (P < 0.001) and fatigue (P < 0.001) were higher than those in group C, whereas the rates of toxicities in group C such as anorexia (P < 0.001) and nausea/vomiting (P < 0.001) were higher than those in group B. The rate of hepatotoxicity in group B was significantly (P < 0.001) higher than that in group C.
Dose reduction was prescribed for 15 patients in the HDI group, whereas 8 patients in chemotherapy group were given dose reduction. Two patients in HDI group and one patient in temozolomide plus cisplatin group were managed by dose interruption. Generally, all the adverse events were mild to moderate in severity and were manageable after dose reduction, dose interruption, or supportive medical intervention. No treatment-related deaths occurred in both groups.
Mucosal melanoma has a very poor prognosis and significantly worse outcomes than cutaneous melanoma, with 5-year disease-free survival rates ranging from 0% to 20% (4–7). Despite the fact that systemic adjuvant therapy has been investigated primarily for cutaneous melanoma, the results observed in that subtype are generally extrapolated to the mucosal melanoma population, considering the rarity of mucosal melanoma in Caucasians (about 1.3% of all melanomas; refs. 1–3). In Asian populations, mucosal melanoma represents the second most common subtype of melanoma (2, 3). Our trial focused on stage II/III mucosal melanoma, aiming to nominate a potential adjuvant therapy for resected mucosal melanoma. Our data suggest that HDI can improve the RFS (median, 9.4 months) and OS (median, 40.4 months) of patients with resected mucosal melanoma, with better results observed for the temozolomide plus cisplatin regimen with respect to RFS (median, 20.8 months) and OS (median, 48.7 months). However, further phase III trials for temozolomide plus cisplatin in Caucasians should be encouraged as our trial is single-centered and the patients are of Chinese population.
In patients with high-risk melanoma, that is, with AJCC stage II (T2–4N0M0) and stage III (TanyN+M0) disease, the rate of disease recurrence ranges between 20% and 60%, with 5-year OS varying between 45% and 70% (36). In our study, the relapse rates in the observation group (100%) as well as the HDI and chemotherapy groups were much higher, which may generally reflect the aggressive characteristics of mucosal melanoma. However, it may be inferred that the relapse rate in the observation group should be less than 100% when it was analyzed in a larger cohort. IFN-α has been the most intensively investigated drug and the only agent currently approved for adjuvant therapy in patients with high-risk melanoma (12–16). In clinical trials of HDI as adjuvant therapy (E1684, E1690, and E1694), very few patients with mucosal melanoma were enrolled, and efficacy for this subset was not specified (13–15). The median RFS of HDI group in current trial was only 9 months, which is much less than the previously reported RFS for cutaneous melanoma and less than our own trial of HDI in acral melanoma (with a median RFS about 22.5 months; refs. 13–15, 25). Considering that all the enrolled patients in current trial were in stage II/III (localized and regional), this difference in RFS may simply reflect the very poor prognosis of mucosal melanoma. However, HDI also seemed to improve the RFS and OS of patients with stage II/III mucosal melanoma as compared with observation, suggesting that HDI may be an adjuvant therapy option for patients with mucosal melanoma. Considering that HDI has not been proven to significantly improve OS of patients with melanoma (12–16), our observations that HDI improved both OS and RFS of mucosal melanoma may indicate that mucosal melanoma may respond better to HDI than other subtypes of melanoma, which may need basic researches to elucidate the underlying molecular, cellular, and histologic differences determining the sensitivity to HDI. In two retrospective reports on anorectal and vaginal melanoma (37, 38), the median PFS was about 23 and 12 months, respectively, which were also different to our data. Therefore, detailed analysis of efficacy of HDI or temozolomide plus cisplatin for Caucasian patients with mucosal melanoma according to stages, anatomic sites, and other factors is required and expected.
Retrospective studies about adjuvant chemotherapy for patients with mucosal melanoma have been reported (22, 23). However interpretable data derived from randomized clinical trials on adjuvant chemotherapy in patients with mucosal melanoma are lacking. In vitro studies suggest that cisplatin may enhance the antitumor activity of temozolomide by its ability to downregulate alkylguanine alkyltransferase (39). Most recently, Flaherty and colleagues reported that biochemotherapy consisting of dacarbazine, cisplatin, vinblastine, interleukin-2 (IL-2), IFN-α, and G-CSF showed significant improvements in RFS (median, 4.0 vs. 1.9 years; P = 0.034) in high-risk stage III melanoma but not OS, which was a coprimary endpoint, as compared with HDI (40), indicating that dacarbazine in combination with cisplatin may be used in adjuvant chemotherapy for melanoma. But most of these studies were retrospective, associated with small sample sizes or were nonrandomized studies, so there is no strong evidence for adjuvant chemotherapy for patients with mucosal melanoma. In our phase II trial, we used the temozolomide plus cisplatin as the chemotherapy arm. Our data suggest that temozolomide plus cisplatin may be advised to patients with mucosal melanoma after surgery as this regimen can significantly improve the RFS [median, 20.8 months; significantly longer than that of HDI in patients with resected mucosal melanoma (P < 0.001) and comparable with the results produced for HDI trials in cutaneous melanoma] and OS (significantly longer than HDI, P = 0.009) of patients with mucosal melanoma, and thus may be considered as a potential regimen for patients with resected mucosal melanoma. The potential mechanisms for the unexpectedly better efficacy of temozolomide plus cisplatin over HDI in mucosal melanoma (dramatically different to what has been reported in cutaneous melanoma) may lie in differences between mucosal melanoma and cutaneous melanoma, such as genetic and metabolic characteristics of the tumor cells, sensitivity to chemotherapy induced apoptosis, and immune environment, which require further basic and clinical investigations. Anyhow, mucosal melanoma is distinctly different from cutaneous melanoma with regard to overall burden of somatic genetic alterations (41, 42).
As compared with previous trials (13–15), the grade 3/4 toxicities, such as leukopenia, hepatotoxicity, and fatigue, in the HDI group was much lower, which was similar to our previous trial of HDI in patients with acral melanoma (25). Moreover, the grade 3/4 toxicities in the temozolomide plus cisplatin group were also not identical to the reports in Caucasians (19, 43, 44). The primary possibility may be the ethnic differences and the differences in melanoma subtypes, with our trial conducted in Chinese population and focusing on mucosal melanoma. For Caucasians, patients with melanoma from different countries may show variable adverse events in response to similar schedules of temozolomide plus cisplatin (19, 43, 44). Therefore, the treatment dosages may be optimized for patients with differential ethnicities or countries once chemotherapy was described as adjuvant therapy.
In sum, our trial has compared temozolomide-based chemotherapy and HDI therapy, two widely selected regimens for adjuvant therapy for melanoma, with surgery alone. Our data suggest that both adjuvant regimens are safe and well tolerated for patients with resected mucosal melanoma, with the temozolomide plus cisplatin regimen more likely to improve RFS and OS than HDI therapy. Our trial is unique in that it explores adjuvant therapy for a specific, uncommon subtype of melanoma. Our trial is the first to suggest that temozolomide-based chemotherapy may be better than HDI at the level of both RFS and OS in this specific subtype of patients with mucosal melanoma. However, as a single-center trial, adjuvant therapy using chemotherapy or HDI in patients with mucosal melanoma may need more randomly controlled trials in the future. In addition, temozolomide, in combination with other agents as ipilimumab, should be tested for efficacy as adjuvant therapy in the future.
Disclosure of Potential Conflicts of Interest
No potential conflicts of interest were disclosed.
Conception and design: B. Lian, L. Si, J. Guo
Acquisition of data (provided animals, acquired and managed patients, provided facilities, etc.): L. Si, L. Mao, S. Li, Y. Kong, B. Tang
Analysis and interpretation of data (e.g., statistical analysis, biostatistics, computational analysis): B. Lian, L. Mao
Writing, review, and/or revision of the manuscript: B. Lian, L. Si
Administrative, technical, or material support (i.e., reporting or organizing data, constructing databases): B. Lian, C. Cui, Z. Chi, X. Sheng
Study supervision: J. Guo
This work was supported by grants from the National Natural Science Foundation of China (81172196, 81102068, and 81272991), Doctoral Fund of Ministry of Education of China (20110001120070 and 20120001110048), the Beijing Nova Program (Z121107002512042), and Major State Basic Research Development Program of China (2013CB911004).
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
The authors thank the staff in the Department of Pathology of our hospital for help in collection and pathological analysis of tissue samples.
Note: Supplementary data for this article are available at Clinical Cancer Research Online (http://clincancerres.aacrjournals.org/).
- Received March 16, 2013.
- Revision received June 5, 2013.
- Accepted June 15, 2013.
- ©2013 American Association for Cancer Research.