| ||||
| ||||
American Society of Clinical Oncology 36th Annual
Meeting
Day 4 - May 23,
2000
Nicholas J. Vogelzang, MD -- Writer: Michelle L. Plante, PharmD
Two of the largest randomized phase III trials in bladder cancer ever conducted were reported at this year's ASCO meeting. These trials have significant implications for treatment of this disease, which takes the lives of 12,000 US citizens annually.
M-VAC vs High-Dose M-VAC
M-VAC (methotrexate, vinblastine, doxorubicin, and cisplatin) is the standard regimen for the treatment of patients with urothelial transitional cell carcinoma (TCC). However, median survival of these patients is only about 1 year. Dr. Cora N. Sternberg,[1] from the Genitourinary Group, European Organization for Research and Treatment of Cancer (EORTC), Brussels, Belgium, presented the results of an EORTC trial examining whether high-dose M-VAC (HD-M-VAC) with growth factor support could improve upon this poor survival. This study was a phase III, randomized, international trial comparing classic M-VAC (n = 129) to HD-M-VAC (n = 134) using granulocyte colony-stimulating factor (G-CSF) support. HD-M-VAC consisted of methotrexate 30 mg/2 on day 1, vinblastine 3 mg/2 on day 2, doxorubicin 30 mg/2 on day 2, and cisplatin 70 mg/2 on day 2, with G-CSF on days 4-11. This cycle was repeated every 14 days. Classic M-VAC was given every 28 days without G-CSF. The median age of the patients enrolled in the study was 61 years. For 81% of the patients in the M-VAC group and 84% of the HD-M-VAC patients, the bladder was the primary tumor site. The median follow-up was 38 months. The results are summarized in Table 1.Table 1. Results of M-VAC vs HD-M-VAC
M-VAC HD-M-VAC P value Median # of cycles (range) 4 (1-8) 6 (1-12) Delivered dose intensity: Methotrexate -- 0.87 Vinblastine -- 0.81 Doxorubicin -- 2.06 Cisplatin -- 2.07 Response rate 58% 73% .008 Complete response 11% 25% .006 Overall survival 25.4% 35.3% NS 2-year PFS 11.6% 24.7% NS Median survival 14.5 months 14.1 months NS
PFS = progression-free survival There was no difference in the incidence of nonhematologic toxicities observed in the 2 groups. However, there was significantly more neutropenia and neutropenic fever in the M-VAC group (26% vs 10%, P < .001). This can be explained by the fact that only 19% of these patients received G-CSF, vs 94% of the HD-M-VAC patients. The toxic death rate was 3% in the M-VAC group vs 5% in the HD-M-VAC group. The results of this trial demonstrate that HD-M-VAC with G-CSF support is feasible and associated with an improvement in the complete response rate. Unfortunately, this did not translate into an improved median survival, but it may have improved the 2-year disease-free survival rate -- the trial was not powered to detect an improvement in that parameter. The hazard ratios suggest an advantage in the HD-M-VAC group in both progression-free and overall survival, but a significantly larger trial is need to confirm these intriguing observations. Meanwhile, all patients on standard dose M-VAC should probably receive G-CSF.
M-VAC vs GC
Because of the significant toxicity and the poor long-term survival (3% to 4% at 6 years) associated with of M-VAC chemotherapy, newer agents are constantly being sought. Three multicenter Phase II trials with gemcitabine and cisplatin (GC) have shown response rates of 35% to 50% against transitional cell carcinoma (TCC). Therefore, the current phase III trial was conducted to compare standard therapy with M-VAC to this new combination. The trial had an 80% power to detect a 4-month improvement in median survival. Dr. von der Masse,[2] from Aarhus University Hospital, Denmark, reported the results of the trial. Patients with locally advanced or metastatic TCC with no prior systemic therapy were randomized to receive either GC (gemcitabine 1000 mg/m2 days 1, 8, and 15 plus cisplatin 70 mg/2 on day 2) or M-VAC, every 28 days. Median follow-up was 19 months. The results are shown in Table 2.Table 2. Results of GC vs M-VAC*
GC
n = 203M-VAC
n = 202Median age (years) 63 63 Complete response 12% 12% Partial response 37% 34% Response rate 50% 46% Time to progressive disease 7.4 months 7.4 months Overall survival 13.8 months 14.8 months
* None of the differences were statistically significant. There were significantly more toxicities in the M-VAC group as shown in Table 3:
GC M-VAC Neutropenia 34% 82% Neutropenic fever 5% 22% Thrombocytopenia 56% 21% Sepsis 1% 12% Death (drug related) 1% 3% Infection 2.5% 15% Mucositis 1% 22% QOL - increased fatigue Hospitalizations 9 admissions/33 days 49 admission/272 days Blood transfusions 0.82 units/cycle 0.68 units/cycle The results of this study show that the combination of gemcitabine and cisplatin is not better than M-VAC in terms of response rate and overall survival. However, it is associated with significantly less toxicity. Given the striking similarity in neutropenic fever rates in this study and the study presented by Sternberg (22% and 26%), the routine use of G-CSF should be strongly considered when M-VAC is used. GC does not require the routine use of G-CSF. These results suggest that gemcitabine and cisplatin can be considered an alternative to M-VAC in the treatment of TCC. The cost and availability of gemcitabine are the main reasons that GC will not completely replace M-VAC. Nevertheless, I plan to use GC as standard therapy in patients with metastatic TCC. An international phase III trial is being planned to compare GC to GC plus paclitaxel. As noted by the discussant, Dr. Howard Scher,[3] of Memorial Sloan-Kettering Cancer Center, New York, NY, there has been no improvement in median survival from metastatic bladder cancer (13.8-15 months), since the observations of Dr. Yagoda in 1985. We still need new active agents in this disease in order to improve the complete response rate and median survival.
Prognosis and Prediction
HER2
The implications of HER2-neu overexpression on prognosis and response to treatment in breast cancer are now becoming well defined. Other types of tumors, such as prostate, ovary, lung, and bladder cancer also overexpress HER2-neu, but the implications of overexpression in bladder tumors are not clear. Dr. Rafael E. Jimenez,[4] of the Barbara Ann Karmanos Cancer Institute in Detroit, Michigan, presented an analysis of HER2-neu expression in 60 cases of metastatic muscle-invasive TCC of the bladder. The aim of the study was to compare the pattern of HER2-neu expression in local and distant metastatic disease to the primary tumor. Tumor samples were stained using the Herceptest and were graded as 0 = absent, 1 = faint, 2 = moderate, and 3 = strong. The sample was considered positive for overexpression if the grade was 2 or 3. The results of the sample analysis are summarized in Table 3.Table 3. HER2 Lymph Node and Metastases Results Based on Primary Tumor Results
HER2 in Primary Tumor HER2 in Regional Lymph Node HER2 in Distant Metastases n (%) Neg (%) Pos (%) Neg (%) Pos (%) Negative 38 (63) 11 (55) 9 (45) 1 (33) 2 (67) Positive 22 (37) 1 (8) 11 (92) 0 (0) 4 (100) Sixty four percent of the primary tumors had heterogeneous staining for HER2-neu. Interestingly, distant metastases were much more likely to overexpress HER2-neu than the primary tumor. Whether this difference reflects overexpression as a necessity for tumor progression or whether this is simply a selection bias within the primary tumor sample is unknown. These results suggest that HER2-neu overexpression in the primary tumor predicts overexpression in regional lymph nodes and distant metastases. The effects of chemotherapy and BCG therapy on intravesical tumor expression of HER2-neu are unknown. Trials using trastuzumab in bladder cancer patients with tumors overexpressing HER2 are now being planned.
Molecular Markers
Dr. Bajorin,[5] of the Memorial Sloan-Kettering Cancer Center in New York, NY, reported a hypothesis-generating study that looked at molecular markers of prognosis in bladder cancer. Bcl-2 and its homologue, bcl-xl, inhibit apoptosis (programmed cell death). Two other members of the family, BAX and bcl-x2, promote apoptosis. p53 mutation is also a recognized prognostic factor. Bajorin used archival material from 111 bladder cancer patients treated with neoadjuvant M-VAC to examine these markers. Eighty-three patients from this series had previously been analyzed to study the implications of p53 overexpression, and some samples had been depleted. Thus, there were only 59 tumors available for this analysis. The median age of the population was 62 years, with 49 males. Forty-seven of the 59 were T3 tumors, and 25% patients were still alive. There was no correlation between molecular end points and clinical features. However, there was a correlation between molecular end points and survival:In a further analysis, patients with wild type p53 (no p53 overexpression) and no bcl-2 overexpression had a 65% 5-year survival. Patients with mutant p53 without bcl-2 overexpression had a 43% 5-year survival, whereas patients with both mutant p53 and bcl-2 overexpression had a 22% 5-year survival. As noted by Dr. Bajorin, the small number of patients in each group and the multiple analyses that had been conducted on these specimens do not allow these data to be more than hypothesis generating. Further studies on molecular prognostic factors are clearly needed. In the discussion session, Dr. Walter Stadler,[6] of the University of Chicago, Illinois, reviewed the clinical and molecular prognostic factors involved in bladder cancer, distinguishing them from predictive factors that forecast which patients will benefit from treatment. The weight of evidence is now strongly supportive of p53 overexpression as being a strong and independent prognostic factor for recurrence and disease-specific survival. E-cadherin, epidermal growth factor receptor and pRb overexpression are potentially important prognostic factors as well. The prognostic significance of Bcl-2 overexpression needs more study, although a recent paper suggests that it may predict for cisplatin sensitivity. In commenting on Jimenez's work, he congratulated this group on sampling metastatic sites, a clinically challenging task, and for emphasizing the heterogeneity of HER2-neu overexpression. No data are available to suggest that HER2-neu overexpression is a prognostic or predictive factor in bladder cancer.
- p53 overexpression, P = .006;
- bcl-2 overexpression, P = .149;
- BAX overexpression, P = .5;
- BAX/bcl-2 ratio, P = .51
Summary
It is clear from the results of the trials presented today that M-VAC is a toxic regimen, and G-CSF should be used prophylactically in all patients receiving this regimen. If G-CSF is used, M-VAC can be given every 2 weeks but is still associated with a 3% to 5% toxic death rate. GC is an attractive alternative to M-VAC in the treatment of bladder cancer, and its lower toxicity profile makes it ideal for elderly patients or patients with comorbidities.
|
