857P - Pitfalls of using the Cockcroft-Gault formula when calculating carboplatin dose for the adjuvant treatment of patients with stage I seminoma

Date 29 September 2012
Event ESMO Congress 2012
Session Poster presentation I
Topics Anti-Cancer Agents & Biologic Therapy
Germ Cell Tumours
Presenter Martin Fehr
Authors M. Fehr1, T. Geldart2, H. Sun3, P.D. Simmonds4, G. Mead4, M. Wheater4, N. Nagaraj5, S. Ellis2, R. von Moos6, R. Cathomas7
  • 1Medical Oncology And Haematology, University Hospital Southampton, SO16 6YD - Southampton/UK
  • 2Medical Oncology, Royal Bournemouth Hospital, BH7 7DW - Bournemouth/UK
  • 3Coordination Center, Swiss Group for Clinical Cancer research (SAKK), 3008 - Bern/CH
  • 4Medical Oncology, University Hospital Southampton, SO16 6YD - Southampton/UK
  • 5Nuclear Medicne, University Hospital Southampton, SO16 6YD - Southampton/UK
  • 6Medical Oncology, Kantonal Hospital Graubünden, Chur/CH
  • 7Medizinische Onkologie, Cantonal Hospital Graub, CH-7000 - Chur/CH



Single dose Carboplatin (C) AUC7 (7 x glomerular filtration rate (GFR) mls/min + 25) is a standard adjuvant treatment option for patients with stage I seminoma. Measuring GFR by a nuclear medicine method represents the gold standard, but is not available in all centres. 24h urine collection may be prone to sampling errors. Estimating GFR by the Cockroft-Gault formula (CG) is popular amongst oncologists but data supporting its use in this situation is limited. Reduction of C dose by 10% in the adjuvant treatment of seminoma stage I is associated with a trend for higher relapse rate.


Data from 202 consecutive, male patients (median age 39, range 23-68 years) with stage I seminoma with complete documentation of GFR by TC99m DTPA, serum-creatinine, age, height and weight were evaluated. Actual C doses (ACD) based on GFR measurement by TC99m DTPA were compared with estimated C doses (ECD) based on GFR estimation using CG formula. Differences between the ACD and ECD were correlated to age and body mass index (BMI) using Fisher's Exact test, Pearson correlation and linear regression.


202 patients were included in the correlation analysis and 181 patients in the comparison of different subgroups, respectively. Lower BMI and higher age were significantly associated with lower ECD, Pearson correlation coefficients 0.59 (p < 0.001) and –0.36 (p < 0.001), respectively. Tables show potential under- and overdosing of C when using CG (ECD expressed as percentage of ACD) in different age groups (Exact test p = 0.001) and groups with different BMI (Exact test p = 0.026)


CG significantly underestimates GFR in leaner and older patients. According to our data over a third of patients with BMI 20–25 or aged 41–50 would be at risk for undertreatment if CG were used routinely. Physicans need to be aware of these limitations when using CG to calculate C dose in patients with stage I seminoma.


All authors have declared no conflicts of interest.

Table: 857P

Age (years) <90% ACD 90-110% ACD >110% ACD BMI (kg/m2) <90% ACD 90-110% ACD >110% ACD
21-30 n = 29 14% 4 72% 21 14% 4 20-25 n = 60 35% 21 57% 34 8% 5
31-40 n = 84 14% 12 68% 57 18% 15 25-30 n = 81 21% 17 69% 56 10% 8
41-50 n = 68 40% 27 56% 38 4% 3 30-40 n = 33 15% 5 58% 19 27% 9