Previous reports have shown that quantification of high tumour grade is of prognostic significance for patients with prostate cancer. In particular, percent Gleason pattern 4 (GP4) has been shown to predict outcome in several studies, although conflicting results have also been reported. A major issue with these studies is that they rely on surrogate markers of outcome rather than patient survival. We have investigated the prognostic predictive value of quantifying GP4 in a series of prostatic biopsies containing Gleason score 3+4=7 and 4+3=7 tumours. It was found that the length of GP4 tumour determined from the measurement of all biopsy cores from a single patient, percent GP4 present and absolute GP4 were all significantly associated with distant progression of tumour, all-cause mortality and cancer-specific mortality over a 10-year follow-up period. Assessment of the relative prognostic significance showed that these parameters outperformed division of cases according to Gleason score (3+4=7 versus 4+3=7). International Society of Urological Pathology (ISUP) Grade Groups currently divide these tumours, according to Gleason grading guidelines, into grade 2 (3+4=7) and grade 3 (4+3=7). Our results indicate that this simple classification results in the loss of important prognostic information. In view of this we would recommend that ISUP Grade Groups 2 and 3 be amalgamated as grade 2 tumour with the percentage of GP4 carcinoma being appended to the final grade, e.g., 3+4=7 carcinoma with 40% pattern 4 tumour would be classified as ISUP Grade Group 2 (40%).