Adjuvant and neoadjuvant therapy: principles and practical considerations

Prof. Eric Winer (Dana-Farber Cancer Institute, Boston, USA) compared the goals, merits, and surgical advantages of adjuvant and neoadjuvant therapy for patients with early breast cancer. He also discussed the clinical meaning of a complete pathological response after neoadjuvant therapy.

The goals of adjuvant and neoadjuvant therapy are somewhat overlapping, Winer demonstrated. For both therapies, eradicating micrometastatic disease and prolonging survival are the primary goals. In addition, the goal of adjuvant therapy is to improve disease-free and/or distant disease-free survival given that these parameters are good surrogates for overall survival or a better quality of life. Additional goals of neoadjuvant therapy are: to decrease the extent of surgery, to provide prognostic information, to identify candidates for additional treatment approaches, and – from a scientific point of view – to test de-escalation strategies and to conduct tissue-intensive trials. Referring to the results of the NSABP B-18 and NSABP B-27 trials, Winer concluded that – so far – there is no difference in (disease-free and/or overall) survival between patients with early breast cancer who had neoadjuvant treatment vs patients who had adjuvant treatment [1]. However, there are circumstances where adjuvant therapy is preferred. For example, if the decision about therapy depends on the anatomic extent of disease (stage 1 Her2-positive disease, T1 triple-negative breast cancer), or if it is impossible to follow the disease (e.g. by imaging or palpation) during neoadjuvant treatment. On the other hand, where it is clear that chemotherapy will be the treatment (e.g. patients with stage 2/3 ER-positive disease) this therapy could as well be given up front, Winer stated. Neoadjuvant therapy is also preferred in case optimal surgery will be facilitated by neoadjuvant treatment. In addition, endocrine neoadjuvant therapy is preferable in patients who are not considered to be candidates for (adjuvant) chemotherapy. However, to successfully administer neoadjuvant therapy an experienced multidisciplinary team is indispensable, Winer cautioned.

One goal of neoadjuvant therapy is to reduce the tumour size. This, theoretically, should increase the number of women who are eligible for (and eventually get) breast-conserving surgery. However, this is not the case, Winer demonstrated with the results of 4 neoadjuvant trials. Although in all studies the experimental neoadjuvant regimens induced a higher pathological complete response (pCR) rate, the rate of breast-conserving surgery only modestly increased [2]. Winer put forward several possible explanations: patient or physician anxiety and bias, multifocal disease at presentation, diffuse calcifications that can not be cleared by conservative surgery, and less than perfect imaging. On the other hand, studies like the ACOSOG Z1071 trial showed that neoadjuvant therapy is able to downstage 40% of cN1 patients to pN0 and that in these patients sentinel node biopsies correctly identified the axillary nodal status ≥90%. This raises the question whether axillary surgery is still necessary after an excellent response to neoadjuvant therapy. This question is now subject of several trials.

In the last part of his lecture, Winer zoomed in on the meaning of a pCR after neoadjuvant therapy. The main question on this topic he addressed was “Is there a consistent correlation between pCR and outcome?” It turned out that this depends on whether we look at the level of clinical trials or on the level of an individual patient. Referring to a review of 12 clinical trials that tested several neoadjuvant regimens in several subtypes of early breast cancer, Winer showed that pCR rates depend both on type of treatment and subset of breast cancer [3]. However, in 3 clinical trials, the magnitude of improvement in pCR did not at all correlate with improvement in event-free or overall survival. Recently, also in the NEO-ALTTO and APHINITY trials, an increase in pCR was not or hardly translated into an increase in survival [4,5]. In contrast to this lack of correlation between pCR and survival in trials, pCR can be a very powerful biomarker in individual patients, Winer illustrated with results from the same review of 12 trials and results from the I-SPY2 trial [6]. Both event-free, distant relapse-free, and overall survival were substantially longer for patients who had a pCR.

Summarised, pCR seems a good predictor for individuals but a poor predictor for the long-term success of a regimen. The explanation for this paradox is, according to Winer, that pCR is related to a better outcome but the relationship between pCR and favourable outcome is not causal. In other words, not all patients with a pCR achieve a favourable outcome and not all patients without a pCR achieve a worse outcome. Therefore, new trials should not aim to increase pCR but instead use pCR on an individual level to test escalation or de-escalation of therapies, Winer concluded.

1. 
   
   1. Rastogi P, et al. J Clin Oncol 2008; 26: 778-785.
   2. King TA, et al. Nat Rev Clin Oncol 2015; 12: 335-343.
   3. Cotezar P, et al. Lancet 2014; 384: 164-172.
   4. Piccart-Gebhart M, et al. J Clin Oncol 2016; 34: 1034-1042.
   5. Von Minckwitz G, et al. N Engl J Med 2017; 377:122-131.
   6. Yee D, et al. SABCS 2017, GS3-08.

Posted on 21 May 201924 March 2021