Purpose: This prospective study was undertaken to evaluate the efficac
y of combination chemotherapy with alternating cycles of vincristine,
doxorubicin, and dexamethasone (VAD) and prednisone, vindesine, carmus
tine, and cyclophosphamide (PECC) in poor-risk multiple myeloma (MM).
Patients and Methods: Forty-four patients were previously untreated; 3
6 had been pretreated with an alkylating agent-containing regimen and
had refractory or relapsed MM. All previously untreated patients had a
high tumor burden at inclusion (stage III according to the Durie and
Salmon classification). Logistic regression and the Cox proportional h
azards models were used to assess the association between patient char
acteristics and response rate and survival, respectively. Results: The
overall response rate was 68% for previously untreated patients, comp
ared with 54% for previously treated patients (P = .16). The median su
rvival time for all patients was 28 months: 53 months in previously un
treated patients, and 18 months in previously treated patients. Univar
iate analysis showed that the predictive factors that had a significan
t affect on survival in the newly diagnosed patients were age, therape
utic response to VAD-PECC, low pretreatment Karnofsky score, high base
line serum beta(2)-microglobulin (beta(2)M) level, bone marrow impairm
ent, and renal insufficiency at the start of treatment. When these par
ameters were used as continuous variables in multivariate analysis, th
ree were found to correlate with survival: serum beta(2)M, followed by
therapeutic response and Karnofsky score. In the previously treated g
roup, only Karnofsky score entered the Cox model. Conclusion: these re
sults indicate that combination VAD-PECC chemotherapy is an effective
treatment that results in high response rates and long-term survival i
n advanced MM.