Paclitaxel and docetaxel are 2 compounds from the new taxoid class of anti-cancer agents. Both drugs are very similar in preclinical activity, mechanism of action and spectrum of clinical activity. Some subtle differences in the intracellular retention of docetaxel may account for its lack of schedule-related myelosuppression and greater potency, and may be relevant to the skin toxicity and oedema which it produces. Early data suggest that there may be differing behaviour of anthracycline/taxoid combinations with respect to cardiotoxicity. Paclitaxel has been studied in several first-line combination therapy trials in ovarian cancer. Here, paclitaxel in combination with a platinum compound seems to have proven itself as a standard regimen. It is uncertain if docetaxel will be evaluated in this context. An abundance of clinical data is available for both analogues in the advanced, metastatic setting of breast cancer. Both also have been compared as single agents with doxorubicin with the results suggesting paclitaxel in a 3-hour infusion is inferior to the anthracycline (in terms of response rate), and those of docetaxel suggesting it is superior to the same dose of doxorubicin. This indirect comparison favours the activity of docetaxel; however, it is clear that in the dose/schedules studied, the taxoid compounds are not equitoxic. Either agent by itself, in the treatment of metastatic breast cancer, remains appropriate; however, lack of cumulative toxicity may make paclitaxel more attractive in some situations where prolonged administration is foreseen. Lung cancer trials have also confirmed the activity of both agents, although docetaxel appears to have slightly more promising activity in previously treated patients than paclitaxel. Paclitaxel in combination with cisplatin has been evaluated in randomised trials as first-line treatment of non-small-cell lung cancer (NSCLC). The results of these trials taken together suggest that this combination has an impact on survival similar to other new regimens now considered 'standard' in the front-line setting in this disease. Unfortunately, despite all the phase II data generated in numerous tumour types, little else can be said about the role of either taxoid in the 'standard' management of malignant disease. It will be some years yet before taxoid-based combinations have been evaluated sufficiently in randomised trials such that the impact of this novel class can be adequately assessed in terms of survival and cure rates.