Analysis of prostate-specific antigen decline as a surrogate for overall survival in metastatic hormone-refractory prostate cancer (HRPC)

5009

Background: A 30% PSA decline following initiation of cytotoxic chemotherapy has been identified as a potential surrogate for overall survival in metastatic HRPC. We sought to examine whether various levels of PSA decline were surrogates for overall survival in TAX327, a randomized trial of q3w docetaxel and prednisone (DP), q1w DP, and q3w mitoxantrone and prednisone (MP). Methods: In this trial of 1006 men with HRPC, 943 men had sufficient data on 3-month post-treatment PSA decline for analysis, and 646 had sufficient data for analysis for a change in PSA kinetics with therapy. Surrogacy was examined for a range of PSA decline from 0% to 90% and PSA normalization, and for post-treatment changes in PSA kinetics. We investigated the Prentice Criteria for surrogacy using Cox proportional hazards models and calculated the proportion of treatment effect explained (PTE) by each surrogate marker. Results: In this analysis, a 30% or greater PSA decline with therapy was identified as the optimal cutoff that correlated with overall survival, based on the highest PTE point estimate (0.66, 95% CI 0.23–1.0), with 1.0 being a perfect surrogate. A 30% decline in PSA in the first 3 months after treatment occurred in 65% of subjects receiving q3w DP, 67% q1w DP, and 44% q3w MP, despite the significant survival benefits seen only with q3w DP. A 30% PSA decline was associated with a hazard ratio (HR) of 0.43 (95% CI 0.36–0.51) for overall survival after adjusting for treatment effect, while treatment effect itself lost significance, indicating surrogacy. Additionally, PSA normalization, changes in PSA kinetics, and pain response were significant prognostic variables, yet were only modest surrogates for the survival benefit seen with q3w DP therapy. Conclusions: In this trial of two schedules of DP as compared to MP for HRPC, a PSA decline of 30% in the first three months following initiation of cytotoxic chemotherapy was found to have the highest degree of surrogacy for overall survival, thus confirming data from the SWOG 9916 trial. However, given that the confidence interval for the estimate of this surrogate effect is wide, overall survival should remain the preferred endpoint for phase III trials of cytotoxic agents in HRPC.

No significant financial relationships to disclose.

A multivariate prognostic nomogram incorporating PSA kinetics in hormone-refractory metastatic prostate cancer (HRPC)

5058

Background: To develop a prognostic model and nomogram using baseline clinical variables to predict death among men with metastatic hormone-refractory prostate cancer (HRPC). Methods: TAX 327 was a clinical trial that randomized 1,006 men with metastatic HRPC to receive 3-weekly or weekly docetaxel or mitoxantrone, each with prednisone. Of these, 635 men had baseline data that included PSA kinetics, with 518 mortality events recorded as of November 2006. We developed a multivariate Cox model and nomogram to predict survival at two, three, and five years. Results: Ten independent prognostic factors were identified in multivariate analysis and include: 1) presence of liver metastases (HR 1.64, p=0.02), 2) number of metastatic sites (HR 1.58 if =2 sites, p=0.001), 3) clinically significant pain at baseline (HR 1.46, p<0.0001), 4) Karnofsky Performance Status (HR 1.42 if =70, p=0.01), 5) type of progression at baseline (HR 1.40 for measurable disease progression and 1.28 for bone scan progression, p=0.002 and 0.014 respectively), 6) pretreatment PSA doubling time (PSADT, HR 1.20 if <55 days, p=0.048), 7) baseline PSA (HR 1.17 per log rise, p<0.0001), 8) tumor grade (HR 1.18 for high grade, p=0.076), 9) alkaline phosphatase (HR 1.26 per log rise, p<0.0001), and 10) hemoglobin (HR 1.10 per unit decline, p=0.006). A PSADT <55 days (median value for this dataset) was associated with other adverse prognostic factors, but was independently associated with shortened overall survival. Men with a PSA less than the median of 114 ng/ml and longer PSADT (=55 days) had a median survival of 24.7 months, while those with higher PSA and shorter PSADT had a median survival of 13.8 months. A nomogram was developed based on this Cox multivariate model and validated internally using bootstrap methods, with a concordance index of 0.69. Conclusions: This multivariate model identified several prognostic factors in men with metastatic HRPC including PSADT, and led to the successful development of a clinically applicable nomogram. External prospective validation may support the wider use of this prognostic baseline model for men with HRPC treated with chemotherapy.

No significant financial relationships to disclose.

Factors affecting pharmacokinetic variability following doxorubicin and docetaxel-based therapy

Current dosing strategies for anticancer drugs result in wide interindividual pharmacokinetic variability. Here, we explored the influence of age, body size, concomitant drugs, dose, infusion duration, and sex on the clearance for doxorubicin and docetaxel in 243 individual patients. Patients received doxorubicin (n=110) or docetaxel (n=152) as monotherapy or in combination chemotherapy regimens. The mean (+/-S.D.) clearance was 63.6+/-22.7 L/h for doxorubicin and 42.8+/-14.9 L/h for docetaxel. Normalisation for body surface area (BSA) reduced the interindividual variability by only <1.7%. Doxorubicin clearance was significantly reduced when administered at doses >50 mg/m(2) or in combination with cyclophosphamide. Upper extremes of body size were associated with increased clearance for both drugs, whereas no consistent effect of age on clearance was discerned. Overall, these findings suggest that incorporation of variables in addition to BSA should be considered in routine dosing strategies for doxorubicin and docetaxel.