The PDQ editorial boards use a ranking system of levels of evidence to help the reader judge the strength of evidence linked to the reported results of a therapeutic strategy. For any given therapy, results of prevention and treatment studies can be ranked on each of the following two scales:
Strength of the study design.
Strength of the endpoints.
Together, the two rankings provide a measure of the overall level of evidence. Screening studies are ranked on strength of study design alone...
Carcinosarcomas (the preferred designation by the World Health Organization) are also referred to as mixed mesodermal or mullerian tumors. Controversy exists about the following issues:
Whether they are true sarcomas.
Whether the sarcomatous elements are actually derived from a common epithelial cell precursor that also gives rise to the usually more abundant adenocarcinomatous elements.
The stromal components of the carcinosarcomas are further characterized by whether they contain homologous elements (such as malignant mesenchymal tissue considered possibly native to the uterus) or heterologous elements (such as striated muscle, cartilage, or bone, which are foreign to the uterus). Carcinosarcomas parallel endometrial cancer in its postmenopausal predominance and in other of its epidemiologic features; increasingly, the treatment of carcinosarcomas is becoming similar to combined modality approaches for endometrial adenocarcinomas.
Patients who present with uterine sarcoma have been treated on a series of phase II studies by the Gynecologic Oncology Group, including the GOG-87B trial, for example.[1,2] These chemotherapy studies have documented some antitumor activity for cisplatin, doxorubicin, and ifosfamide. These studies have also documented differences in response leading to separate trials for patients with carcinosarcomas and leiomyosarcomas. As an example, in patients previously untreated with chemotherapy, ifosfamide had a 32.2% response rate in patients with carcinosarcomas  and a 17.2% partial response rate in patients with leiomyosarcomas.
A randomized comparison that was seen in the GOG-108 trial, for example, of ifosfamide with or without cisplatin for first-line therapy for patients with measurable advanced or recurrent carcinosarcomas demonstrated a higher response rate (54% vs. 34%) and longer progression-free survival (PFS) on the combination arm (6 months vs. 4 months), but there was no significant improvement in survival (9 months vs. 8 months).[Level of evidence: 1iiA] The follow-up GOG-0161 [NCT00003128] study utilized 3-day ifosfamide regimens (instead of the more toxic 5-day regimen in the preceding study) for the control and for a combination with paclitaxel (with filgrastim starting on day 4). The combination was superior in response rates (45% vs. 29%), PFS (8.4 months vs. 5.8 months), and overall survival (13.5 months and 8.4 months). The hazard ratio for death favored the combination 0.69 (95% confidence interval, 0.49–0.97).[Level of evidence: 1iiA] In this study, 52% of 179 evaluable patients had recurrent disease, 18% had stage III disease, and 30% had stage IV disease. In addition, imbalances were present in the sites of disease and in the use of prior radiation therapy, and 30 patients were excluded for wrong pathology.