Preoperative systemic treatment for operable disease: biological correlates of response to neoadjuvant treatment
- GN Hortobagyi1
© BioMed Central Ltd 2007
Received: 23 May 2007
Published: 19 June 2007
Experiments on preclinical models have suggested that preoperative introduction of chemotherapy of endocrine therapy can improve survival by abrogating a postsurgical growth spurt and by addressing micrometastatic disease at an earlier stage than when given postoperatively. Clinical trials have not confirmed this hypothesis, although they have demonstrated that preoperative systemic therapy reduced the overall tumor burden, expanded the indications for breast-conserving surgery to patients with more advanced disease, and provided an in vivo assessment of response that facilitated the safe and effective administration of systemic treatment. Preoperative administration of systemic therapy also provides an improved investigational model, since sequential monitoring with the primary tumor in situ offers the opportunity for multiple biopsies to monitor the biological effects of treatment. Large randomized trials have shown preoperative chemotherapy to be at least equivalent in disease-free and overall survival to the same chemotherapy administered postoperatively. Emerging data suggest, however, that the effects of systemic therapy in general, and preoperative therapy in particular, vary with different subclasses of breast cancer (BC). The first such observations were based on the estrogen receptor (ER). It is now widely accepted that the magnitude of benefit from chemotherapy is proportionately more modest for patients with ER+ tumors than for ER- tumors. This is dramatically expressed in preoperative trials, where the pathological complete remission (pCR) rate is fourfold to sixfold higher for ER- BC than for ER+ BC. Similar variation in the pCR rate is observed by grade. More recent reports have indicated that the pCR rate was very low for invasive lobular cancers (almost always ER+) compared with invasive ductal cancers. Paradoxically, although pCR identifies a group of patients with improved survival compared with patients who do not achieve pCR, patient groups with lower pCR rates (those with ER+ tumors, those with low grade or lobular cancer) have better overall survival than those who tend to have higher pCR rates. This observation emphasizes the importance of understanding the biological heterogeneity of BC, and parallels the observations made in lymphoma over the past several decades: high-grade lymphomas respond more readily to chemotherapy but have lower survival rates in early follow-up than low-grade lymphomas. Studies based on gene expression profiling in BC have confirmed the existence of at least three distinct forms of BC: ER+/HER2-, HER2+, and 'triple-negative'. These three groups differ by much more than the individual gene (ER or HER2) expression, and their clinical course and responsiveness to different treatments is quite different too. Thus, HER2+ and triple-negative tumors achieve a high pCR rate (40–50%) with standard combinations, while ER+/HER2- tumors do not (pCR rate <10%). There are other ramifications to drug sensitivity, and, most importantly, gene profiling leads to the identification of potential new targets for therapeutic intervention. Validation of such novel targets and development of specific therapeutics might be the best legacy of neoadjuvant treatment.