Weekly chemotherapy with cisplatin, vincristine, doxorubicin, and etoposide is an effective treatment for advanced thymic carcinoma




Thymic carcinoma is a rare neoplasm that often disseminates or metastasizes. The role of chemotherapy in treating this malignancy is unclear. The purpose of the current study was to determine the efficacy and tolerability of a weekly chemotherapy regimen consisting of cisplatin, vincristine, doxorubicin, and etoposide (CODE) for the treatment of advanced thymic carcinoma.


The authors retrospectively reviewed 18 patients with thymic carcinoma who were treated between 1996 and 2002. Twelve of these patients had unresectable advanced diseases and received weekly chemotherapy according to the CODE regimen. The CODE regimen consisted of cisplatin (25 mg/m2, intravenously [i.v.]; weekly administration), vincristine (1 mg/m2, i.v.; administered during Weeks 1, 2, 4, 6, and 8), doxorubicin (40 mg/m2, i.v.; administered during Weeks 1, 3, 5, 7, and 9), and etoposide (80 mg/m2, i.v.; administered for 3 days during Weeks 1, 3, 5, 7, and 9).


The responses of all 12 patients to the CODE regimen were assessed. A partial response was achieved in 5 patients, and the overall response rate was 42%. Only one patient experienced disease progression. The median progression-free survival period was 5.6 months (range, 2–39 months). The overall survival period ranged from 6 to 79 months, with a median survival period of 46 months. Based on the Kaplan–Meier method, the estimated 1-year and 2-year survival rates were 80% and 58%, respectively. The most common side effects were hematologic toxicities, and only mild nonhematologic toxicities were experienced.


Weekly chemotherapy treatments according to the CODE regimen were effective and tolerated by patients with advanced thymic carcinoma. Cancer 2003;98:926–31. © 2003 American Cancer Society.

DOI 10.1002/cncr.11606

Thymic carcinoma is recognized as a distinct entity from thymoma in thymic epithelial tumors.1–4 However, this tumor is uncommon. Approximately 700 cases have been reported worldwide since Shimosato et al.5 first reported a series of 8 squamous cell carcinomas of the thymus in 1977.

Thymic carcinoma is characterized histologically as a thymic epithelial tumor showing clear-cut cytologic atypia and cytoarchitectural features that are not specific to the thymus. A variety of histopathologic subtypes have been described.6 These subtypes include epidermoid keratinizing carcinoma, epidermoid nonkeratinizing carcinoma, lymphoepithelioma-like carcinoma, sarcomatoid carcinoma, clear cell carcinoma, basaloid carcinoma, mucoepidermoid carcinoma, papillary carcinoma, and undifferentiated carcinoma.

A standard treatment for thymic carcinoma has not been established, but the treatment of first choice is often surgical resection. According to the literature, however, approximately 70% of thymic carcinomas have invaded the surrounding organs at the time of diagnosis and approximately 30% have disseminated or metastasized.3, 7 Therefore, many thymic carcinomas are unresectable when first discovered. The prognosis of patients with thymic carcinoma is generally poor compared with the prognosis of patients with thymoma.3, 8 Chemotherapy or chemoradiotherapy often is performed in patients with unresectable advanced thymic carcinoma, but the role of chemotherapy has remained unclear because of the rarity of this tumor. Although the reported chemotherapeutic agents used with thymic carcinoma include cisplatin, doxorubicin, and etoposide, the total published experience consists of several case reports and a few small series.9–12

We have encountered 18 cases of thymic carcinoma during the last 7 years at the National Cancer Center Hospital East (Kashiwa, Japan). In the current study, 12 patients who received a weekly chemotherapy regimen of cisplatin, vincristine, doxorubicin, and etoposide (CODE) were selected. The efficacy and tolerability of this regimen for the treatment of thymic carcinoma were analyzed retrospectively.


We reviewed 50 thymic epithelial tumors in the database of the Division of Thoracic Oncology, National Cancer Center Hospital East (Kashiwa, Japan) that were treated at this institution between 1996 and 2002. Eighteen of these tumors were histologically proven to be thymic carcinomas. Of these 18 patients, 5 were treated using surgical resection and 1 received radiotherapy alone. The remaining 12 patients were treated using chemotherapy because of local invasion and/or distant metastasis at the time of presentation. All 12 patients were treated using the weekly CODE regimen.

A physical examination, a complete blood cell count, serum electrolyte evaluation, blood chemistry panel, chest radiography, computed tomography (CT) scans of the thorax and abdomen, a bone scan, and a magnetic resonance imaging scan of the brain were performed for all patients before the start of chemotherapy.

Every treatment cycle was repeated at 1-week intervals. If the toxicities were acceptable and the tumor responded to the treatment, patients were expected to complete a maximum of nine cycles of chemotherapy. In brief, the CODE regimen consists of a weekly dose of cisplatin (25 mg/m2 intravenously [i.v.]), vincristine (1 mg/m2 i.v.) during Weeks 1, 2, 4, 6, and 8, doxorubicin (40 mg/m2 i.v.) during Weeks 1, 3, 5, 7, and 9, and etoposide (80 mg/m2 i.v.) for 3 days during Weeks 1, 3, 5, 7, and 9. Prophylactic recombinant human granulocyte—colony-stimulating factor (G-CSF; 50 μg/m2) was administered subcutaneously on the days when the cytotoxic drugs were not given. Treatment was delayed 1 week if the leukocyte count was less than 2000/μL or the platelet count was less than 50,000/μL. Granisetron was administered weekly on Day 1 as an antiemetic. Prophylactic antibiotics were not administered routinely. Toxicities were assessed retrospectively using the patients' medical records and graded according to the National Cancer Institute Common Toxicity Criteria, version 2.0. The response to chemotherapy was evaluated using the World Health Organization (WHO) standard response criteria.13 The survival period was measured from the first day of the CODE regimen to the date of death or the date of the last follow-up. The progression-free survival period was measured from the first day of the CODE regimen to the documentation of progression or death. The survival curves were calculated using the Kaplan–Meier method.14


Patient Characteristics

Patient characteristics are listed in Table 1. The median patient age was 63 years (range, 35–69 years). Seven patients were men. Eight patients had chest pain, coughing, sputum, hoarseness, or dyspnea at the time of presentation. Four patients were asymptomatic. Their tumors were detected incidentally on chest radiography examinations performed during annual health check-ups or during follow-up examinations for other diseases. None of the patients had paraneoplastic syndromes, such as myasthenia gravis or pure red cell aplasia. The serum levels of carcinoembryonic antigen and squamous cell antigen tumor markers were within the normal range for most patients. According to the clinical staging system described by Masaoka et al.,15 one patient had Stage III disease, seven patients had Stage IVA disease, and four patients had Stage IVB disease. Eleven patients had pleural and/or pericardial dissemination and 4 patients had distant metastases, such as in the lung or bone. When the WHO histologic classification for thymic tumors was applied to the current series,6 five patients had epidermoid nonkeratinizing carcinoma (Fig. 1). The tumors in the remaining seven patients were diagnosed as unspecified carcinomas because of the small pathologic specimens obtained by needle biopsy. However, when the WHO criteria of thymic carcinoma (defined as the absence of a primary tumor at sites other than the anterior mediastinum, either at the time of presentation or at follow-up) were applied to our series, the tumors were diagnosed definitely as thymic carcinoma and lung carcinoma was ruled out.

Table 1. Patient Characteristics
Patient no.Age (yrs)GenderHistologyMasaoka stageSites of direct invasion and/or disseminationDistant metastasis site(s)
  1. F: female; M: male; epidermoid: epidermoid nonkeratinizing carcinoma; Unspecified: unspecified carcinoma; SVC: superior vena cava.

267FUnspecifiedIVaSternum, rib, pleura
455MEpidermoidIVaPleura, pericardium
548MUnspecifiedIVaPleura, pericardium
667FUnspecifiedIVbPleura, pericardiumLung, Bone
758MUnspecifiedIVaPleura, pericardium
863MEpidermoidIVbPleura, pericardiumLung
969MUnspecifiedIVaLeft atrium, pericardium
1060FEpidermoidIVbPleura, pericardiumLung
1169MEpidermoidIVaSVC, pleura
1235FUnspecifiedIVaPleura, pericardium
Figure 1.

Thymic carcinoma, epidermoid nonkeratinizing subtype. The tumor is composed of cells exhibiting distinct cytologic atypia but lacking overt keratinization (Patient 11).

Treatment Outcomes

Eleven patients received six or more cycles of chemotherapy; the median was seven cycles for the entire group (Table 2). One patient received only three chemotherapy cycles because of disease progression (Patient 5).

Table 2. Treatment Outcomes of Patients with Advanced Thymic Carcinoma
Patient no.No. of chemotherapy cyclesResponseAdjuvant therapyRecurrencePFS (mos)Survival (mos)
  1. PFS: progression-free survival; PR: partial response; NC: no change; PD: progressive disease.

19PRSurgeryLocal3979 (died)
26PRRadiotherapyLocal1417 (died)
39PRLocal, lung642 (alive)
49NCPleural effusion546 (died)
53PDLocal2 7 (died)
69PRLocal, lung813 (died)
77NCLocal212 (died)
86NCLocal2632 (alive)
96PRRadiotherapyNone1616 (alive)
107NCPleural effusion913 (alive)
119NCLung3 6 (alive)
128NCLocal5 6 (alive)

The responses of all 12 patients were assessed. Patients were evaluated for response using CT scans after every three to four cycles of chemotherapy. Table 2 shows the treatment outcome of each patient. A partial response (PR) was achieved in five patients, six patients exhibited no change, and only one patient demonstrated progressive disease (PD). The overall response rate was 42% (95% confidence interval, 15.2–72.3%). The median progression-free survival period was 5.6 months (range, 2–39 months; Fig. 2). Local disease recurrence was observed in six patients, and local disease recurrence plus metastasis was observed in five patients. Three of the five patients with a PR received adjuvant therapy after the completion of chemotherapy. Two of these patients received thoracic radiotherapy and one patient underwent a surgical resection. One patient with a PR experienced disease recurrence 8 months after the completion of chemotherapy (Patient 6). This patient was re-treated using the same regimen and exhibited a minor response.

Figure 2.

Progression-free survival curve for 12 patients with advanced thymic carcinoma. The median progression-free survival period was 5.6 months (range, 2–39 months).

The median follow-up time was 14 months (range, 6–42 months). Figure 3 shows the overall survival curve. The median survival time was 46 months (range, 6–79 months). Based on the Kaplan–Meier method, the estimated 1-year and 2-year survival rates were 80% and 58%, respectively. To date, six patients have died of this malignancy.

Figure 3.

Overall survival curve for 12 patients with advanced thymic carcinoma. Based on the Kaplan–Meier method, the estimated 1-year and 2-year survival rates were 80% and 58%, respectively.

The toxicities of the CODE regimen are shown in Table 3. Hematologic toxicities were observed most frequently. Five patients had Grade 4 leukopenia, and seven patients had Grade 4 neutropenia. Nine patients experienced Grade 3 or 4 anemia, and four patients experienced Grade 3 thrombocytopenia. Two patients required prophylactic platelet transfusions but did not experience any bleeding episodes. Four patients showed Grade 3 febrile neutropenia, but treatment-related death was not observed. The most common nonhematologic toxicities were nausea and vomiting. Only one patient experienced Grade 3 nausea and hyponatremia, but these symptoms improved for a short period after the completion of chemotherapy. Overall, the nonhematologic toxicities were regarded as mild.

Table 3. Toxicity
ToxicityNCI-CTC grade (no. of patients)
  1. NCI-CTC: National Cancer Institute Common Toxicity Criteria.

Febrile neutropenia40


Thymic carcinoma is a rare neoplasm that accounts for 5–36% of all thymic epithelial tumors.1, 3, 16 Thymic carcinomas usually are more advanced at the time of presentation and have a poorer prognosis than do thymomas.8 Several Phase II chemotherapy trials for advanced thymomas have demonstrated that these tumors are relatively sensitive to chemotherapy.17–20 In contrast, clinical trials for thymic carcinoma are difficult to conduct because of the rarity of this tumor. To our knowledge, a Phase II chemotherapy trial for advanced thymic carcinomas only has not been performed.

A review of the medical literature produced four small series that evaluated the efficacy of platinum-based chemotherapy for the treatment of thymic carcinoma (Table 4).21–24 Oshita et al.21 reported three PRs among seven patients with advanced thymic carcinoma who received the chemotherapy regimen of cisplatin, doxorubicin, cyclophosphamide, and etoposide (PACE). Loehrer et al.22 reported the results of another study that was performed as part of an intergroup trial in the United States. Although thymic carcinoma is considered to be distinct from thymoma, both tumors were regarded as the same thymic epithelial malignancy in this study. The VIP chemotherapy regimen, consisting of etoposide, ifosfamide, and cisplatin, was used in the intergroup trial. Of the eight patients with thymic carcinoma, only two patients had a PR. The 2-year estimated survival rate was 42% for patients with thymic carcinoma. Lucchi et al.23 reported four complete responses and three PRs in seven patients with Masaoka Stage III thymic carcinoma. In their study, the patients were treated with cisplatin, epirubicin, and etoposide as neoadjuvant chemotherapy. After the completion of chemotherapy, all the patients underwent a surgical resection and postoperative radiotherapy. They suggested that a multimodality treatment might improve the outcome of patients with thymic carcinoma. Koizumi et al.24 reported that six PRs were achieved in eight patients with thymic carcinoma, including two patients with small cell carcinoma, after treatment with ADOC chemotherapy. The ADOC chemotherapy consisted of cisplatin, doxorubicin, vincristine, and cyclophosphamide. In these 4 reports describing the use of platinum-based chemotherapy for the treatment of thymic carcinoma, a combined total of 14 of 30 patients (46%) exhibited an objective response to the chemotherapy regimen. We speculate that thymic carcinomas may be moderately sensitive to platinum-based chemotherapy.

Table 4. Studies on the Use of Chemotherapy for the Treatment of Thymic Carcinoma
StudyNo. of patientsChemotherapy regimenOutcome
  1. CDDP: cisplatin; ADM: doxorubicin; ETOP: etoposide; CPA: cyclophosphamide; EPI: epirubicin; IFO: ifosfamide; VCR: vincristine; PR: partial response; CR: complete response.

Oshita et al., 1995217CDDP, ADM, ETOP, CPA3 PR
Loehrer et al., 2001228CDDP, ETOP, IFO2 PR
Lucchi et al., 2001237CDDP, ETOP, EPI3 PR and 4 CR (only Masaoka Stage III)
Koizumi et al., 2003248CDDP, ADM, VCR, CPA6 PR (included 2 small cell carcinomas)
Yoh et al., 2003 (current study)12CDDP, VCR, ADM, ETOP5 PR

We analyzed the efficacy of the CODE regimen for 12 patients with unresectable advanced thymic carcinoma. This regimen is a well designed, intensive weekly chemotherapy treatment that has been reported to be effective for the treatment of small cell lung carcinoma (SCLC).25 The four drugs have different mechanisms of action and their toxicities do not completely overlap. In the current study, the CODE regimen produced an overall response rate of 42% against thymic carcinoma. Although further follow-up is required, the estimated 2-year survival rate was 58%. We included a larger number of patients compared with previous studies. We believe that the CODE regimen may be an effective treatment for patients with unresectable advanced thymic carcinoma.

The main side effects of the CODE regimen were hematologic toxicities. Although G-CSF was administered routinely, Grade 4 leukopenia and neutropenia were observed in 42% and 58% of patients, respectively. Grade 3–4 anemia and thrombocytopenia were observed in 75% and 33% of patients, respectively. However, the frequencies of these hematologic toxicities were approximately equal to that of previously reported PACE21 or VIP22 regimens. To reduce further the neutropenia of the CODE regimen, the dose modification criteria described by Murray et al.25 for the treatment of SCLC should be followed. Nonhematologic toxicities were mild and transient in all patients. All toxicities were easily manageable and no treatment-related deaths occurred.

In conclusion, treatment with the CODE regimen produced a significant response in patients with advanced thymic carcinoma and should be considered as a useful treatment for this condition. However, the results of the current study must be confirmed by prospective Phase II trials. A multicenter Phase II trial of the CODE regimen for the treatment of thymomas, but not thymic carcinomas, currently is being conducted by the Japanese Clinical Oncology Group.