Primary systemic treatment of operable breast cancer with targeted agents

Primary systemic treatment (PST) is the standard therapy for inflammatory and locally advanced breast cancer but it is currently more and more used even in some subsets of patients with operable disease, instead of conventional adjuvant treatments. So far, this strategy has not yet produced a clear survival advantage, as compared to standard postoperative therapy.1 However, PST can allow for breast conservative surgery when up-front mastectomy would be recommended, or can offer a better cosmetic result in case of unfavourable breast-to-tumor size ratio, without jeopardizing survival2,3. Moreover, PST permits an in vivo evaluation of treatment efficacy according to therapy-induced changes of the nodule, and allows to identify subgroups of patients with different prognosis: the patients who achieve a pathological complete response (pCR) benefit most from the treatment and have an excellent prognosis while those with residual breast and/or nodal disease after PST have a worse prognosis. To date, several tumor characteristics have been identified as predictors of the probability of pCR: poorly differentiated tumors, with high proliferation rate HT negative or HER2 positive are more likely to respond to chemotherapy.4-7 However, even if the pCR rate is different across the breast molecular subtypes, once a pCR is achieved, prognosis is good irrespectively of tumor phenotype.7 As pCR rate is now considered a powerful surrogate of long-term disease-free survival, several studies have been conducted to improve the rate of pCR and the more promising results have been obtained with sequential anthracyclines and taxanes. In the ECTO trial, 1355 patients have been randomized to receive post-operative doxorubicin times four followed by CMF times four (A CMF) or post-operative doxorubicin plus paclitaxel times four followed by CMF times four (AP CMF) or four courses of pre-operative AP followed by four courses of CMF. The rate of breast conservative surgery was 65% in patients randomized in the preoperative arm as comparing to 34% in patients not treated upfront with chemotherapy (p>0.001); a pCR in breast and axillary nodes was observed in 20% of the cases.8 In the study conducted by the Aberdeen Breast Group, 162 patients received four courses of preoperative chemotherapy with cyclophosphamide, doxorubicin, vincristine and prednisone (CVAP). Patients achieving a partial or a complete response were randomised to either four further cycles of CVAP or four courses of docetaxel, while non-responders received four cycles of docetaxel. In this relatively small trial, it was shown that both cCR (34% versus 62%) and pCR (18% versus 34%) were in favour of the docetaxel arm.9 In the GEPARTRIO study, 286 patients were evaluated after two courses of preoperative docetaxel, doxorubicin and cyclophosphamide (TAC): non responders were randomized to four cycles of TAC or four courses of capecitabine and navelbine, while responders received four additional courses of TAC. The pCR rate was 22.3% in responding patients receiving six courses of TAC, 7.3% in non responding patients receiving four additional courses of TAC and 3.1% only in those patients randomized to four courses of capecitabine-navelbine.10 In the NSABP-B27, 2411 patients were randomized to receive: four courses of AC followed by surgery; four courses of AC followed by four courses of docetaxel and then surgery; four courses of AC followed by surgery followed by four courses of post-operative docetaxel. It was shown that the addition of docetaxel almost doubled the pCR rate (26.1% versus 12.9% and 14.4% in the two pre-operative AC arms).11 Another interesting trial conducted at the M.D. Anderson Cancer Center has evaluated the sequential administration of weekly vs 3-weekly paclitaxel, followed by FAC: the administration of paclitaxel on a weekly schedule resulted in a significantly higher pCR rate (28.2% versus 15.7%, p=0.02).12 Furthermore, in the AGO trial, dose dense sequential epirubicin followed paclitaxel produced a significant DFS and OS improvement as compared to three weekly epirubicin-paclitaxel combination. The pCR rate was 18% in the dose intensified arm vs 10% in the standard treatment arm (p=0.0045). The 5 year DFS rates were 70% versus 59% (p=0.011) and the 5 year OS rates were 83% versus 77% (p=0.041) in the dose intensified vs the standard arm respectively.13 Furthermore PST represents the ideal setting where new targeted agents can be tested: 1) a molecular response can be evaluated a few hours or days after the first drug dose; 2) a metabolic response can be measured by FDG-PET after a few weeks of therapy; 3) treatment activity can be measured as objective response after few months; 4) at surgery, after the end of preoperative therapy, a pathological response can be defined and tumor biomarkers can be characterized again. There are now many successful examples of preoperative treatments with targeted agents for operable disease. In the MDACC trial for instance, combining trastuzumab to chemotherapy in HER2 positive patient resulted in more than doubled pCR rate (66.7% versus 25%) as compared to chemotherapy alone.14 The phase III NOAH trial is the largest neoadjuvant study that evaluated the addition of trastuzumab to an anthracyclines- and taxanes-based chemotherapy for patients with HER2-positive locally advanced breast cancer (LABC): 327 patients were randomized to receive three cycles of doxorubicin (60 mg/m) and paclitaxel (150 mg/m) q3w, 4 cycles of paclitaxel (175 mg/m q3w) and 3 cycles of CMF (cyclophosphamide 600 mg/m, methotrexate 40 mg/m, 5-fluorouracil 600 mg/m q4w) on days 1 and 8, with or without concomitant H (8 mg/kg loading dose then 6 mg/kg q3w for 1 year) before surgery. In parallel, LABC patients screened as HER2-negative received the same regimen of chemotherapy. The primary endpoint was event-free survival (EFS), secondary endpoints were pCR, ORR, OS and safety. EFS rate at 3 years was significantly better in the chemotherapy + trastuzumab arm compared with chemotherapy alone: 70.1% versus 53.3%, respectively (HR 0.56; p=0.007); moreover, both ORR and pCR were significantly higher in the chemotherapy + trastuzumab arm compared to chemotherapy alone: 89% versus 77% for ORR, respectively (p=0.02); 39% versus 20% for pCR, respectively (p=0.002). This analysis of the NOAH study established neoadjuvant chemotherapy + trastuzumab as a standard treatment option in women with HER2-positive LABC.15 Nowadays the combination of neoadjuvant chemotherapy and trastuzumab is becoming a powerful treatment for all HER2 positive breast cancer, even in case of operable disease. In this setting, we are conducting a phase II randomized trial (CHER-LOB trial) in which patients with HER2 positive tumors larger than 2 cm are randomized to receive: chemotherapy plus trastuzumab (arm A), chemotherapy plus lapatinib (arm B), or chemotherapy plus trastuzumab and lapatinib (arm C). Chemotherapy consists in: paclitaxel 80 mg/m2 weekly for 12 weeks, followed by four courses of FEC (5FU 600 mg/m2 + epirubicin 75 mg/m2 + cyclophosphamide 600 mg/m2 iv) q21; trastuzumab is administered at 2 mg/kg weekly; following the second safety report of the Independent Data Monitoring Committee, due to the occurrence of grade 3 diarrhoea in 20% of the patients randomized to arm B and in 41% of the patients randomized to arm C, lapatinib is administered at 1250 mg po daily in arm B, and at 750 mg po daily in arm C. Primary endpoint of this study is the percentage of pCR defined as complete disappearance of invasive tumor cells in both breast and axillary nodes. Secondary aims: breast objective response, breast conservative surgery, safety, molecular responses, gene expression related to pCR. The updated activity and safety data of CHER-LOB trial have been recently reported at the 2008 San Antonio Breast Cancer Symposium.16 In the preoperative setting of hormone receptor positive breast cancer, the administration of endocrine agents is becoming a more and more attractive alternative, too. Up to now, preoperative endocrine therapy has been limited to elderly patients with locally advanced disease unfit for chemotherapy. Neoadjuvant hormonal therapy is traditionally considered less attractive due to the very low rate of pCR achievable with this treatment modality and because of the optimal duration of endocrine treatments is till five years. Nevertheless, preoperative endocrine treatments are now increasingly used for the following reasons: 1) chemotherapy can induce a lower percentage of pCR in hormone receptor positive tumors thus suggesting a major role for endocrine manipulation; 2) even if the pCR rate with endocrine agents is very low, the majority of hormone receptor positive tumors shrink after few months, thus allowing for more conservative surgery; 3) evaluation of tumor biomarkers (i.e. Ki 67) permits to identify very early those patients more likely to benefit from treatment; 4) preoperative trials allow to identify more active endocrine agents in a more limited time span, with reduced number of patients. For instance, in the IMPACT trial 330 postmenopausal patients have been randomized to receive preoperative anastrozole or tamoxifen or the association of both drugs: no differences were observed in terms of overall responses and breast conserving surgery. However, a greater inhibition of tumor proliferation was achieved in the anastrozole arm as compared to tamoxifen or to the combination: the Ki-67 levels after two weeks of therapy seem an independent predictive factor for relapse free survival.17-18 On the contrary, Eiermann et al. observed a significantly superior overall objective response rates and breast conservative surgery in the letrozole group as compared to the tamoxifen group in 337 postmenopausal patients.19 However, several patients show a primary or an acquired resistance to hormones, suggesting the role of cross-talk between hormonal receptors and other intracellular pathways as mechanism of resistance to endocrine therapy: thus, blocking both the pathways by combining endocrine and target therapies can enhance the antitumor activity, even in the neoadjuvant setting. For example, in a randomized trial conducted in women with hormone receptor and EGFR positive tumors, gefitinib monotherapy induced a tumor shrinkage in 54% of the patients, while the combination gefitinib + anastrozole vs. gefitinib alone resulted in a higher inhibition of tumor proliferation.20 Everolimus (RAD001) has recently demonstrated to significantly increase letrozole efficacy in neoadjuvant therapy of 270 patients randomly assigned to four months of letrozole and either everolimus (10 mg/day) or placebo: response rate by clinical palpation in the everolimus arm was higher than that with letrozole alone (68.1% versus 59.1%). A reduction in Ki-67 expression after 15 days of treatment occurred in 57% of patients in the everolimus arm and in 30% of patients in the placebo arm (p<0.01).21 The LET-LOB trial is a double blind phase II study in which hormone receptor positive, Her2 negative, postmenopausal patients are randomized to receive letrozole plus placebo or letrozole plus lapatinib for 6 months. Primary endpoint of the study is the clinical objective response rate, and secondary endpoints are the percentage of pCR, the safety profile of the combination of hormonal treatment plus lapatinib and the evaluation of the inhibition of intermediate and final biomarkers of the proliferative and the apoptosis pathways induced by the hormonal treatment plus lapatinib or placebo. Preliminary activity and safety data of this trial have been recently updated at 2009 ASCO Annual Meeting.22 In conclusion, PST has an established role in locally advanced disease, can increase the rate of breast conserving surgery in operable disease and can be safely proposed to any patient for whom adjuvant treatment can be recommended on the basis of pre-surgical tumor stage and biology. Moreover, this treatment modality allows to identify subsets of patients with poor prognosis after primary therapy eligible for trials with new agents. Finally, the opportunity to collect tumor samples before and after therapy permits to observe the treatment effects on tumor biology in a relatively short time, thus speeding the development of targeted agents.