Paclitaxel has emerged as an important agent in the treatment of breast cancer. The efficacy and tolerability of this agent, as well as its lack of cross-resistance with anthracyclines, have led to its inclusion in combination treatments . Taxane- and anthracycline-containing regimens have been extensively used, and one prospective phase III clinical study reported a higher response rate and a longer overall survival for paclitaxel-doxorubicin compared to 5-fluorouracil, doxorubicin and cyclophosphamide . These results, however, have been challenged by other authors who did not demonstrate a survival advantage or higher toxicity for taxane-based therapy [4–8]. In view of these inconsistent results, attempts have been made to find schedules that are capable of increasing efficacy without worsening toxicity .
Gemcitabine has been shown to be effective and safe as a single agent and in combination regimens and, given its different mechanism of action and partial non cross-resistance with anthracyclines and taxanes, represents an ideal candidate for a three-drug regimen. Such a combination (gemcitabine, epirubicin and paclitaxel – GET) was recently tested in a clinical setting. After very promising results in a phase II study, with a 92% response rate , the GET regimen failed to demonstrate a higher efficacy than the FEC (5-fluorouracil, epirubicin and cyclophosphamide) combination in a multicenter randomized phase III trial. In addition to the lack of a significant advantage of GET over FEC in terms of response rate and time to progression, the GET arm showed higher toxicity .
We designed a clinical protocol based on a preclinical schedule that had produced the highest synergistic drug interaction in an attempt to improve the clinical efficacy obtained with empirically designed treatment, or at least to achieve similar results but with lower drug dosages and milder toxicity [8, 9].
In the present study, we demonstrated that the addition of gemcitabine to doxorubicin and paclitaxel produced clinical results using low doses of the three drugs, with consequently minimal non hematological toxicity. In particular, the lack of cardio- and neurotoxicity is an important advantage because it permits anthracycline and taxane re-treatment. Whilst it is also clear that gemcitabine significantly increased myelosuppression, this toxicity was manageable with primary or secondary G-CSF prophylaxis. It is important to stress that the toxic death in our population occurred before the first administration of gemcitabine, i.e., after one cycle of doxorubicin and paclitaxel given at lower doses than usual.
Activity data appear comparable with those reported in the previously mentioned phase III randomized trials of anthracycline-paclitaxel based chemotherapy in metastatic breast cancer. In these trials [3–9], response rates ranged from 46% to 68% and median time to progression varied from 6 to 9.8 months. The experimentally designed GAT regimen showed a fairly similar efficacy profile in terms of response rate and time to progression, and results were particularly interesting for the small group of stage IIIB patients, who all responded to treatment and the majority of whom were operable.
Difficulties in translating in vitro results into clinical practice are inevitable. In our preclinical study , the synergistic sequence of doxorubicin → paclitaxel → gemcitabine was defined in two cancer cell lines, BRC-230 and MCF7, characterized by a 100% growth fraction and a doubling time of around 30 hours. The cell kinetics of human breast cancer are considerably different in that the growth fraction is remarkably low and it takes several days for the cell population to double. Moreover, we know that in clinical practice delays may occur in drug administration due to toxicity, patient compliance, or logistic problems, and these timing violations can affect clinical outcome. These and other important issues about translational research remain unsolved and must be addressed in clinical research.