The purpose of this study is to determine the overall best tumor response rate to dacarbazine given until disease progression as assessed by RECIST criteria, CT and clinical exams in patients with metastatic sarcomas.

The two reasons why dacarbazine was eliminated from treatment options for patients with metastatic sarcoma included inability to effectively address the drug's major toxicities (emesis and neutropenia) and the prevailing opinion that the drug was less effective than other chemotherapeutic agents.

Specific Aim #1: The primary endpoint of this study is to determine the overall best tumor anatomic response rate (CR- complete response, PR - partial response, SD - stable disease, or PD - progressive disease) to dacarbazine given until disease progression as assessed by RECIST criteria using CT and clinical examination in patients with metastatic sarcoma.

The prevailing opinion that dacarbazine was less effective than other chemotherapeutic agents in this setting was not based on data from controlled randomized clinical trials. Indeed, we are not aware of a single randomized trial that was conducted and reported which compared the anti-tumor activity of single agent dacarbazine to that of other active single agents in this patient population. However, phase two trials clearly established that dacarbazine had anti-tumor activity in the treatment of metastatic sarcoma, and our personal experience at this institution has confirmed that this is true. In addition, randomized trials demonstrated that the addition of dacarbazine to doxorubicin increased tumor response rates over doxorubicin alone.12 The literature supports the conclusion that dacarbazine has anti-tumor activity in the treatment of metastatic sarcoma.

Historically, in most studies, tumor response to dacarbazine was assessed by WHO criteria. However, current assessment of tumor response usually is based on RECIST criteria, which differ from the WHO criteria, as shown:

Studies have not been performed to determine the tumor anatomic response rate of single agent dacarbazine using RECIST criteria. This study will determine the tumor anatomic response rate as assessed by RECIST criteria to dacarbazine in patients with metastatic sarcoma. Modern methods of CT scans will be used to assess tumor response which contrasts with the earlier methods to assess tumor response to dacarbazine used in most of the published reports. These methods included physical examination and conventional X-ray or first generation, lower resolution CT scans. The current methods of radiologic assessment of tumor response are superior to those used over 15 years ago. The latest generation of CT scans more accurately measure and image a tumor mass, which may better assess tumor response to therapy.

Specific Aim #2: To determine the overall risk of nausea/emesis (any grade) and neutropenia (grade 3 or 4) with dacarbazine when given with current antiemetic agents (5-hydroxytryptamine-3 serotonin antagonist, dexamethasone, and aprepitant) and with pegfilgrastim.

Historically, dacarbazine-induced toxicities such as emesis and myelosuppression were common and led to significant dose reductions and delays which could have negatively impacted the drug's anti-tumor activity. When dacarbazine was initially identified as a potentially effective agent for the treatment of sarcoma, the anti-emetic drugs available had limited efficacy. Also, there were no measures available to prevent chemotherapy-induced neutropenia. Today, there are very effective drugs that can prevent and reduce the frequency of both chemotherapy-induced emesis and neutropenia.

Dacarbazine carries the risk of emesis (all grades) of >90% of cases when administered without antiemetics13. A three-drug anti-emetic combination of a 5-hydroxytryptamine-3 serotonin antagonist, dexamethasone, and aprepitant is the current recommendation from ASCO when administering highly emetogenic chemotherapy drugs13. Hesketh, et al reported that the risk of emesis (all grades) following highly emetogenic chemotherapy (cisplatin) and pre-medication with this three drug anti-emetic regimen was 27% compared to a two drug regimen of ondansetron and dexamethasone in which the risk was 48% (p< 0.001)14.

Prior studies showed that the risk of grade 3 or 4 neutropenia following dacarbazine given without granulocyte- colony stimulating factors was 36%15. Granulocyte- colony stimulating factors (pegfilgrastim or neupogen) are effective agents in preventing chemotherapy-induced neutropenia. Crawford, et al showed in a randomized trial that risk of chemotherapy- induced grade four neutropenia was one day with G-CSF compared to six days with placebo15. Subsequent randomized trials showed equivalent efficacy of pegfilgrastim compared with neupogen16. Vogel, et al reported in a phase three double blinded randomized trial of patients with breast cancer receiving docetaxel 100 mg/m2 that pegfilgrastim compared to placebo reduced the incidence of febrile neutropenia ( 1% vs 17%, p< 0.001) and febrile neutropenia-related hospitalization (1% vs 14%, p< 0.001)17.

In this trial, we hypothesize that the implementation of these newer anti-nausea agents (5-hydroxytryptamine-3 serotonin antagonist, dexamethasone, and aprepitant) and granulocyte-colony stimulating factor (pegfilgrastim) as a primary prophylactic strategy will reduce the frequency of dacarbazine-induced nausea/emesis (any grade) and neutropenia (grade 3 or 4).

Specific Aim #3: To compare the SUV at up to three target tumor sites and to determine the overall tumor metabolic response (complete metabolic response, partial metabolic response, stable metabolic disease or progressive metabolic disease (CMR, PMR, SMD, or PMD) as assessed by FDG-PET/CT performed at baseline and then after every three cycles of treatment with dacarbazine.

Studies have not been performed to determine the changes in FDG uptake by PET imaging following dacarbazine in patients with metastatic sarcoma. There is limited published data about changes in FDG uptake by PET imaging following other chemotherapy agents in patients with metastatic sarcoma. However, there is an emerging body of data showing the prognostic impact of early tumor response to targeted agents (imatinib or sunitinib) as assessed by FDG-PET in patients with metastatic GIST18. In this trial, we will determine the tumor metabolic response to dacarbazine as assessed by FDG-PET/CT and correlate this to the tumor anatomic response as assessed by CT scans.

Inclusion Criteria:

• Histologically proven diagnosis of soft tissue or bone sarcoma
• Metastatic or locally recurrent and unresectable sarcoma which progressed after one or more prior chemotherapy regimens (excluding adjuvant chemotherapy).
• At least one measurable tumor lesion (by CT scan) At least one FDG avid (SUV ≥ 3) tumor lesion (by PET/CT) which must have been performed at this institution. At least one of these target lesions must be ≥ 1.5 cm in smallest dimension as measured on the baseline CT
• Age greater than 18 yrs old
• ECOG Performance Status of 0-2
• Baseline ANC ≥ 1000/uL, Hgb ≥ 8 Gr/dL, platelets ≥ 100,000/ dL.
• Baseline serum creatinine </= 2.0 mg/dL
• Baseline serum total bilirubin </= 2.0, AST or ALT < 3x ULN
• No active infection
• Signed Informed Consent by patient or legally authorized representative

Exclusion Criteria:

• Current pregnancy or breast feeding.
• A serious uncontrolled medical disorder that in the opinion of the Investigator would impair the ability of the subject to receive protocol therapy.
• Chemotherapy, radiation therapy, or investigational agents given with the last 21 days.
• Investigational agents given with the last 30 days
• Uncontrolled diabetes mellitus. (Subjects with a fasting blood glucose > 200 at time of PET scanning may need to reschedule to another day after consulting with appropriate physicians.)