The objective of this work was to propose and demonstrate a novel technique for the assessment of tumour pharmacokinetic parameters together with a regionally estimated vascular input function. A breast cancer patient T2*-weighted dynamic contrast enhanced MRI (DCE-MRI) dataset acquired at high temporal resolution during the first-pass bolus perfusion was used for testing the technique. Extraction of the lesion volume transfer constant K-trans together with the intravascular plasma volume fraction nu(p) was achieved by optimizing a capillary input function with a measure of cardiac output using the principle of intravascular indicator dilution theory. For a region of interest drawn within the breast lesion a nu(p) of 0.16 and a K-trans of 0.70 min(-1) were estimated. Despite the value of nu(p) being higher than expected, estimated K-trans was in accordance with the literature values. In conclusion, the technique proposed here, has the main advantage of allowing the estimation of breast tumour pharmacokinetic parameters from first-pass perfusion T2*-weighted DCE-MRI data without the need of measuring an arterial input function. The technique may also have applicability to T1-weighted DCE-MRI data.