Radiotherapy versus carboplatin for stage I seminoma: Updated analysis of the MRC/EORTC randomized trial (ISRCTN27163214)

1 Background: At ASCO 2004 we reported a randomised trial (MRC TE19, EORTC 30982) comparing 1 course carboplatin (C) at AUC×7 with adjuvant radiotherapy (RT) for stage I seminoma. With median follow-up (FU) of 4 yrs, C was shown to be non-inferior for relapse free rate (1° outcome) and there was a reduction in contralateral testicular gem cell cancers (GCT). Because late relapses and 2nd GCT can occur beyond 5 and 10 yrs we have continued FU of these patients (pts). We report the updated results, and further analysis of C dose and outcome in relation to method of assessing renal function. Methods: The 1° randomisation was RT versus one injection of C dosed at 7× (GFR+25) based on EDTA(n=357) and 90% of this dose if based on creatinine clearance(n=202). The trial was powered to exclude absolute differences in the 2 yr relapse-free rates (RFR) of > 3%. RFR were determined by the Kaplan-Meier method, and hazard ratios (HR) from the Cox regression model; HR>1 favors RT. Results: Between 1996–2001, 1447 pts were randomised in a 3:5 ratio (C=573, RT=904). Median FU is now 6.5 yrs with documented minimum 5 yr FU on 1148 pts. RFRs at 5 yrs are 95% (C) and 96% (RT) (HR 1.25, 90% CI 0.83, 1.89); an increase in 5yr RFR of >3.6% can be excluded with 95% confidence. Only one death from seminoma (RT) has been reported. There was a significant difference in the rate of new GCTs (2 on C vs 15 on RT), giving a HR of 0.22 (95% CI 0.05, 0.95 p=0.03). High levels of pre-treatment FSH (>12 iu/L) were associated with an increased risk of developing a 2nd GCT (HR: 8.57 (95% CI 1.82 –40.38)) Analysing the variation in dose received in the C group, showed that those who received at least 99% of the 7AUC dose (n=347) ignoring GFR method had 5 yr RFR of 96.1% (95% CI 93.4–97.7) compared with 92.6% (88.0–95.5) in those who received lower doses (n=212, HR: 0.51 (0.24 - 1.07) p=0.08)). Analysis of the available data on pathological risk factors demonstrated that larger tumours (>4cm) had poorer RFR (HR: 3.68 (1.49 - 9.13)). Conclusion: With prolonged FU this trial confirms the non-inferiority of single dose C (AUC7) vs RT in terms of RFR but with a reduced risk of 2nd GCT in the C arm. The observation on the impact of C dosage, though not significant in its own right, reinforces the safety of adequately dosed C and may support studies of higher dose C. Author Disclosure Employment or Leadership Consultant or Advisory Role Stock Ownership Honoraria Research Expert Testimony Other Remuneration Medical Research Council (non-commercial sponsor)