Anthracyclines and Taxanes in Breast Cancer Treatment
As opposed to the evaluation of the best sequence of adjuvant hormone therapy, something that has been undertaken by a randomized trial with nearly 8000 women and long-term end points, studies to date assessing the role of reverse sequences of anthracycline- and taxane-based regimens have been relatively small. A large retrospective study that analyzed close to 1600 patients showed that the usual adjuvant sequence in which an anthracycline was followed by a taxane was associated with a significantly higher risk of death (but not of relapse), when compared with the reverse sequence. The collection of adjuvant trials described herein, with a combined total of nearly 350 patients, showed that the RDI for the taxane was higher, and there were fewer dose reductions when a taxane was administered first. In the neoadjuvant setting, over 1400 patients were analyzed retrospectively and nearly 1300 patients were enrolled in clinical trials, which collectively showed similar or increased pCR rates for sequences in which the taxane was administered first, in comparison with the reverse sequence. The largest randomized phase III trial confirmed the results of the smaller phase II studies, with a statistically and clinically significant difference in pCR rate favoring the taxane-first sequence (20% versus 15%). After neoadjuvant chemotherapy, pCR is consistently higher in hormone-receptor-negative tumors, when compared with hormone-receptor-positive counterparts. Most studies published to date that compared reversed chemotherapy sequences did not stratify according to hormone-receptor status, and potential imbalances may influence their results. The same goes for HER-2 testing, which was not carried out in most studies. In the pre-trastuzumab era, Stearns et al. showed in a small group of HER-2-positive tumors that the likelihood of achieving a clinical complete response was higher after anthracyclines than after taxanes. A recent trial that added trastuzumab to chemotherapy showed very high pCR rates irrespective of drug scheduling. Therefore, active anti-HER-2 treatments may overcome the impact related to the sequence of chemotherapy agents in this subset of breast tumors.
In essence, none of the trials to date, either in the adjuvant or neoadjuvant setting, has shown disadvantages in terms of efficacy or toxicity for sequences in which the taxane was administered first; the side-effect profile in such trials was related to each specific agent rather than to the sequence used.
The biological rationale for administering taxanes before anthracyclines rests on limited preclinical and clinical observations. Guo et al. assessed the capacity of paclitaxel or doxorubicin to induce cross-resistance to each other and to additional chemotherapeutic agents using cell lines resistant to one of these agents. The selection of MCF-7 breast tumor cells for resistance to doxorubicin generated a population of cells which were >4000-fold cross-resistant to paclitaxel, whereas an identical selection of cells for resistance to the taxane generated a population of cells with only slight (4-fold) cross-resistance to the anthracycline. The exact mechanism for this difference is not completely understood, but seems to be associated with the expression of P-glycoprotein and breast cancer resistance protein. In the study by Taghian et al., the administration of paclitaxel first led to significant differences in IFP and tumor oxygenation, regardless of tumor size or response measured by ultrasound. Such findings raise the hypothesis that tumors undergoing such changes could ultimately have a better overall response because of the improved penetration of the second drug and the improved cytotoxicity of oxygen-dependent drugs (such as doxorubicin and cyclophosphamide). Another potential mechanism refers to senescence, a response to nonlethal stress that results in persistent cytostasis with a distinct morphological and biochemical phenotype; such response is a usual consequence of DNA-damaging agents. Thus, it is possible that doxorubicin-induced senescence may render cells resistant to further therapies, as opposed to the initial treatment with less potent inducers of senescence, such as the taxanes. Finally, a biological observation stemming from an unpublished study indicates a third differential effect of treatment with anthracyclines and taxanes: the analysis of the release of circulating tumor cells after neoadjuvant therapy suggested that longer survival might be achieved by using a taxane followed by anthracycline, compared with the reverse sequence.
In the metastatic setting, only a few trials have been designed with the aim of comparing reverse anthracycline and taxane sequences as their primary or secondary objectives. The largest randomized trial to compare drug sequencing was Intergroup trial E1193, in which 739 patients were assigned to receive either single-agent paclitaxel, single-agent doxorubicin, or their combination in the first line; at the time of disease progression, there was a protocol-specified cross-over to the alternative drug in the single-agent arms, which allowed for the secondary comparison of both sequences. The delivery of paclitaxel followed by doxorubicin upon disease progression resulted in no difference in terms of response rate, time-to-treatment failure, or overall survival, when compared with the reverse sequence. French investigators randomized 136 women to one of three arms for total treatment durations of eight cycles: docetaxel alternated with FEC; docetaxel followed by FEC; or the reverse sequence (FEC followed by docetaxel). The response rate was lower (although not statistically significant so) in the arm with docetaxel followed by FEC, when compared with the two other arms in which docetaxel was administered first. The median time to progression and the 2-year survival rates were similar in the three treatment groups. The authors suggested further investigation for the two regimens with higher response rates. In a smaller randomized trial, the comparison of sequences was a primary aim, as patients received either three cycles of docetaxel followed by three cycles of doxorubicin and cyclophosphamide, or the reverse sequence. Accrual for this study was suspended after 33 patients had been randomized, based on statistical considerations pertaining to trial design. Although median survival times were 2.5 years in the docetaxel-first arm and 1.1 year in the anthracycline-first arm, such difference was not statistically significant, and progression-free survival and objective response rates were similar in both arms. Finally, the sequence with the taxane first appeared slightly less toxic in a randomized phase I trial that compared escalating doses of epirubicin and paclitaxel versus the reverse sequence. It is unknown whether the outcomes of these two trials can be translated into the adjuvant setting.
The combination of biological, clinical, and demographic features, as well as the genomic and transcriptomic characteristics of tumors may better tailor the treatment of the individual patient with early breast cancer. Although anthracyclines have been used for decades, validated biomarkers that are able to predict benefit from these agents are currently lacking. Despite encouraging results from some prospective studies, several candidate genes have been assessed in studies that were retrospective or confounded by the concurrent administration of other chemotherapeutic agents, thus making it difficult to derive clear correlations between genotype and the effect of anthracyclines. Likewise, the contribution of pharmacogenomics to individualizing taxane therapy in breast cancer has been limited to date, and the full potential of gene profiling in early breast cancer is yet to be explored.
Since the personalized use of specific cytotoxic agents for the (neo)adjuvant chemotherapy of breast cancer remains a clinical challenge, the decision to administer anthracyclines and taxanes in these settings is currently based on classical biological, clinical, and demographic features that are associated with the risks of recurrence. Most adjuvant and neoadjuvant trials have typically incorporated a taxane after the anthracycline-based regimen on the basis of historical evolution, although some recent studies have used the taxane before the anthracycline. To our knowledge, there is no planned or published phase III trial assessing the relative merit of reverse sequences of anthracyclines and taxanes in the adjuvant or neoadjuvant settings. As pointed out by Wildiers et al., it may be difficult to secure funding for a large phase III trial investigating this issue. There are only two registered ongoing phase II trials that compare drug sequencing, and they may provide further information in the adjuvant and neoadjuvant settings, albeit with little power for long-term outcomes. Therefore, given the available information discussed; herein, we believe that a taxane followed by an anthracycline is a sequence option that can be incorporated into daily clinical practice.
Discussion
As opposed to the evaluation of the best sequence of adjuvant hormone therapy, something that has been undertaken by a randomized trial with nearly 8000 women and long-term end points, studies to date assessing the role of reverse sequences of anthracycline- and taxane-based regimens have been relatively small. A large retrospective study that analyzed close to 1600 patients showed that the usual adjuvant sequence in which an anthracycline was followed by a taxane was associated with a significantly higher risk of death (but not of relapse), when compared with the reverse sequence. The collection of adjuvant trials described herein, with a combined total of nearly 350 patients, showed that the RDI for the taxane was higher, and there were fewer dose reductions when a taxane was administered first. In the neoadjuvant setting, over 1400 patients were analyzed retrospectively and nearly 1300 patients were enrolled in clinical trials, which collectively showed similar or increased pCR rates for sequences in which the taxane was administered first, in comparison with the reverse sequence. The largest randomized phase III trial confirmed the results of the smaller phase II studies, with a statistically and clinically significant difference in pCR rate favoring the taxane-first sequence (20% versus 15%). After neoadjuvant chemotherapy, pCR is consistently higher in hormone-receptor-negative tumors, when compared with hormone-receptor-positive counterparts. Most studies published to date that compared reversed chemotherapy sequences did not stratify according to hormone-receptor status, and potential imbalances may influence their results. The same goes for HER-2 testing, which was not carried out in most studies. In the pre-trastuzumab era, Stearns et al. showed in a small group of HER-2-positive tumors that the likelihood of achieving a clinical complete response was higher after anthracyclines than after taxanes. A recent trial that added trastuzumab to chemotherapy showed very high pCR rates irrespective of drug scheduling. Therefore, active anti-HER-2 treatments may overcome the impact related to the sequence of chemotherapy agents in this subset of breast tumors.
In essence, none of the trials to date, either in the adjuvant or neoadjuvant setting, has shown disadvantages in terms of efficacy or toxicity for sequences in which the taxane was administered first; the side-effect profile in such trials was related to each specific agent rather than to the sequence used.
The biological rationale for administering taxanes before anthracyclines rests on limited preclinical and clinical observations. Guo et al. assessed the capacity of paclitaxel or doxorubicin to induce cross-resistance to each other and to additional chemotherapeutic agents using cell lines resistant to one of these agents. The selection of MCF-7 breast tumor cells for resistance to doxorubicin generated a population of cells which were >4000-fold cross-resistant to paclitaxel, whereas an identical selection of cells for resistance to the taxane generated a population of cells with only slight (4-fold) cross-resistance to the anthracycline. The exact mechanism for this difference is not completely understood, but seems to be associated with the expression of P-glycoprotein and breast cancer resistance protein. In the study by Taghian et al., the administration of paclitaxel first led to significant differences in IFP and tumor oxygenation, regardless of tumor size or response measured by ultrasound. Such findings raise the hypothesis that tumors undergoing such changes could ultimately have a better overall response because of the improved penetration of the second drug and the improved cytotoxicity of oxygen-dependent drugs (such as doxorubicin and cyclophosphamide). Another potential mechanism refers to senescence, a response to nonlethal stress that results in persistent cytostasis with a distinct morphological and biochemical phenotype; such response is a usual consequence of DNA-damaging agents. Thus, it is possible that doxorubicin-induced senescence may render cells resistant to further therapies, as opposed to the initial treatment with less potent inducers of senescence, such as the taxanes. Finally, a biological observation stemming from an unpublished study indicates a third differential effect of treatment with anthracyclines and taxanes: the analysis of the release of circulating tumor cells after neoadjuvant therapy suggested that longer survival might be achieved by using a taxane followed by anthracycline, compared with the reverse sequence.
In the metastatic setting, only a few trials have been designed with the aim of comparing reverse anthracycline and taxane sequences as their primary or secondary objectives. The largest randomized trial to compare drug sequencing was Intergroup trial E1193, in which 739 patients were assigned to receive either single-agent paclitaxel, single-agent doxorubicin, or their combination in the first line; at the time of disease progression, there was a protocol-specified cross-over to the alternative drug in the single-agent arms, which allowed for the secondary comparison of both sequences. The delivery of paclitaxel followed by doxorubicin upon disease progression resulted in no difference in terms of response rate, time-to-treatment failure, or overall survival, when compared with the reverse sequence. French investigators randomized 136 women to one of three arms for total treatment durations of eight cycles: docetaxel alternated with FEC; docetaxel followed by FEC; or the reverse sequence (FEC followed by docetaxel). The response rate was lower (although not statistically significant so) in the arm with docetaxel followed by FEC, when compared with the two other arms in which docetaxel was administered first. The median time to progression and the 2-year survival rates were similar in the three treatment groups. The authors suggested further investigation for the two regimens with higher response rates. In a smaller randomized trial, the comparison of sequences was a primary aim, as patients received either three cycles of docetaxel followed by three cycles of doxorubicin and cyclophosphamide, or the reverse sequence. Accrual for this study was suspended after 33 patients had been randomized, based on statistical considerations pertaining to trial design. Although median survival times were 2.5 years in the docetaxel-first arm and 1.1 year in the anthracycline-first arm, such difference was not statistically significant, and progression-free survival and objective response rates were similar in both arms. Finally, the sequence with the taxane first appeared slightly less toxic in a randomized phase I trial that compared escalating doses of epirubicin and paclitaxel versus the reverse sequence. It is unknown whether the outcomes of these two trials can be translated into the adjuvant setting.
The combination of biological, clinical, and demographic features, as well as the genomic and transcriptomic characteristics of tumors may better tailor the treatment of the individual patient with early breast cancer. Although anthracyclines have been used for decades, validated biomarkers that are able to predict benefit from these agents are currently lacking. Despite encouraging results from some prospective studies, several candidate genes have been assessed in studies that were retrospective or confounded by the concurrent administration of other chemotherapeutic agents, thus making it difficult to derive clear correlations between genotype and the effect of anthracyclines. Likewise, the contribution of pharmacogenomics to individualizing taxane therapy in breast cancer has been limited to date, and the full potential of gene profiling in early breast cancer is yet to be explored.
Since the personalized use of specific cytotoxic agents for the (neo)adjuvant chemotherapy of breast cancer remains a clinical challenge, the decision to administer anthracyclines and taxanes in these settings is currently based on classical biological, clinical, and demographic features that are associated with the risks of recurrence. Most adjuvant and neoadjuvant trials have typically incorporated a taxane after the anthracycline-based regimen on the basis of historical evolution, although some recent studies have used the taxane before the anthracycline. To our knowledge, there is no planned or published phase III trial assessing the relative merit of reverse sequences of anthracyclines and taxanes in the adjuvant or neoadjuvant settings. As pointed out by Wildiers et al., it may be difficult to secure funding for a large phase III trial investigating this issue. There are only two registered ongoing phase II trials that compare drug sequencing, and they may provide further information in the adjuvant and neoadjuvant settings, albeit with little power for long-term outcomes. Therefore, given the available information discussed; herein, we believe that a taxane followed by an anthracycline is a sequence option that can be incorporated into daily clinical practice.
Source...