OBJECTIVES To compare chemoradiation with 100mg/m(2) cisplatin every three weeks to 20mg/m(2) on five days every four weeks for locally advanced squamous cell carcinoma of the head-and-neck (LASCCHN). MATERIALS AND METHODS In 230 patients receiving chemoradiation for LASCCHN, 100mg/m(2) cisplatin every three weeks (N=126) and 20mg/m(2) cisplatin on five days every four weeks (N=104) were retrospectively compared. Chemoradiation plus eleven characteristics (T-/N-classification, performance score, gender, age, tumor site, grading, surgery, radiation technique, pre-chemoradiation hemoglobin, cumulative cisplatin dose) were analyzed for locoregional control (LRC), metastases-free survival (MFS) and overall survival (OS). Chemoradiation groups were compared for adverse events. RESULTS On univariate analyses, chemoradiation had no impact on LRC (p=0.53), MFS (p=0.67) and OS (p=0.14). On multivariate analysis of LRC, T-classification (p=0.045) and hemoglobin (p<0.001) were significant. On multivariate analysis of MFS, performance score (p=0.028) was significant. On multivariate analysis of OS, performance score (p=0.009) and hemoglobin levels (p=0.002) achieved significance. Chemoradiation with 100mg/m(2) cisplatin was associated with more pneumonia/sepsis (p=0.003), grade ⩾2nausea/vomiting (p<0.001), grade ⩾2 nephrotoxicity (p=0.005), grade ⩾2 xerostomia (p=0.002), grade ⩾3 hematotoxicity (p=0.052) and grade ⩾2 ototoxicity (p=0.048). CONCLUDING STATEMENT 20mg/m(2) cisplatin on five days every four weeks was associated with fewer adverse events than 100mg/m(2) cisplatin every three weeks. 100mg/m(2) cisplatin was not significantly superior to 20mg/m(2) cisplatin regarding LRC, MFS and OS. Given the limitations of a retrospective study, 20mg/m(2) cisplatin appeared preferable. The results should be confirmed in a randomized trial.