Imaging performance in guiding response to neoadjuvant therapy according to breast cancer subtypes: A systematic literature review

Highlights

• 15 observational studies examined imaging performance for measuring response during NAC.

• Evidence to conclude on the preferred imaging technique per BC subtype is lacking.

• ¹⁸FDG-PET/CT seems to be preferred in ER-negative/HER2positive BC over MRI.

• MRI and ¹⁸FDG-PET/CT are promising in both triple negative and HER2-positive BC.

• Consensus should be reached on: interval time, cut-off values, and pCR definitions.

Abstract

Monitoring therapeutic response to neoadjuvant chemotherapy(NAC) is likely to improve NAC effectiveness in breast cancer(BC). Imaging performance seems to vary per tumour subtype(by ER and HER2 status), therefore we performed a systematic review on subtype specific imaging performance in monitoring NAC in BC.

Studies examining imaging performance in predicting pathologic complete response(pCR) during NAC in BC subtypes were selected. Per study, negative- and positive predictive value, sensitivity(se) and specificity(sp), AUC and accuracy were derived.

Fifteen/106 articles were included. Inter-study variability was revealed in: monitoring interval, response and pCR definitions. In ER-positive/HER2-negative BC, ¹⁸1F FDG-PET/CT showed se/sp of 38%–89%/74%–100%, MRI showed se/sp of 35%–37%/87%–89%. In triple negative BC, ¹⁸1F FDG-PET/CT showed se/sp of 0%–79%/95%–100%. ¹⁸1F FDG-PET/CT showed in ER-positive/HER2-positive BC se/sp of 59%/80% and in ER-negative/HER2-positive 27%/88%.

Evidence on imaging performance in monitoring NAC according BC subtypes is lacking. Consensus should be reached in: definitions of pCR, response and monitoring interval before starting well-designed studies.

Introduction

In 2012, 1.7 million new cases of breast cancer were diagnosed worldwide. Despite research and improvements in breast cancer treatment, breast cancer is still: one of the most prevalent cancers overall, the most prevalent cancer among women, and one of the main causes of death (WHO, 2012). Research on new treatment approaches is thus of evident interest.

Neoadjuvant chemotherapy(NAC) showed to be at least equally effective as adjuvant chemotherapy (Mauri et al., 2005) while having additional advantages (Fisher et al., 1997, van der Hage et al., 2001), such as the ability to monitor therapeutic response during treatment (Gralow et al., 2008). Early therapeutic response assessed by imaging seems to be a predictor of pathologic complete response(pCR) (Marinovich et al., 2012), usually defined as absence of any residual invasive tumour cells in the original tumour bed and axilla (Kaufmann et al., 2006). PCR itself predicts long-term survival, especially in HER2-positive and triple negative(TN) tumours (Chollet et al., 2002, Von Minckwitz et al., 2012), monitoring early therapeutic response may be used to guide systemic treatment, which is called a response-guided NAC approach (von Minckwitz et al., 2013). Under this scenario, patients could be monitored after a specific number of NAC cycles, and according to their response at imaging, their further systematic treatment could be tailored, i.e. responders continue with the same initial treatment, and non-responders can be switched to a presumably non-cross-resistant regimen(Fig. 1) (von Minckwitz et al., 2013).

Currently, there is no definite guideline to assess response to NAC during treatment. Previous authors proposed physical examination plus mammography and ultrasound, but their performance seems to be limited (Yeh et al., 2005, Hamisa et al., 2015, Londero et al., 2004). Therefore, performance examination of more advanced techniques, i.e. magnetic resonance imaging(MRI) and PET—Computed Tomography(PET/CT) is of interest. So far, meta-analyses have shown sensitivities and specificities of 68% and 91% for dynamic contrast-enhanced(DCE)-MRI (Wu et al., 2012), 93% and 82% for diffusion-weighted(DW)-MRI (Wu et al., 2012) and 84% and 71% for ¹⁸F FDG-PET/CT (Cheng et al., 2012) respectively. On the basis of these findings, MRI is currently the technique mainly used in clinical practice. While these techniques seem to already have better performance, recent studies have shown that breast cancer subtype affects imaging performance (Loo et al., 2011, Hayashi et al., 2013, Ko et al., 2013). Hence, personalizing the use of imaging techniques based on subtypes may further improve their performance in evaluating therapeutic response (Loo et al., 2011, Humbert et al., 2012).

As there is no subtype-specific guidance on imaging techniques to monitor therapeutic response during NAC to guide in further treatment regimen, this paper aims to create an overview of current knowledge on the performance of imaging techniques in breast cancer subtypes based on expression of ER and HER2.