Evaluation of treatment response and resistance in metastatic renal cell cancer (mRCC) using integrated ¹⁸F–Fluorodeoxyglucose (¹⁸F–FDG) positron emission tomography/magnetic resonance imaging (PET/MRI); The REMAP study

Systemic treatment options for metastatic RCC have shifted to targeted therapies in the last decade following demonstration of improved progression free survival in Phase III clinical trials [6‐12]. Renal cell carcinoma can be a highly vascular tumour, for example, related to Von-Hippel Lindau gene mutations and accumulation of hypoxia inducible factor (HIF). Transcriptional targets of HIF include vascular endothelial growth factor (VEGF) and platelet derived growth factor (PDGF), with upregulation of angiogenesis, hence why anti-angiogenic targeted therapies have been successful in metastatic RCC.

A number of targeted agents have received FDA approval for metastatic RCC. These include the small molecule multi-kinase inhibitors including pazopinib, sunitinib, sorafenib and axitinib, the mammalian target of rapamycin (mTOR) inhibitors everolimus and temsirolimus, and the monoclonal antibody Bevacizumab, an inhibitor of vascular endothelial growth factor A. These agents selectively inhibit neo-angiogenesis and cell proliferation respectively through a variety of signalling pathways [13] including the RAS-RAF-MEK-ERK and PI3K-AKT-mTOR pathways (Fig. 1). These agents have been shown to positively impact survival with transient stabilisation of disease in 70% of patients as well as providing scope for several lines of treatment [6‐12]. More recently clinical trials of immunotherapeutic agents have also shown promise. These agents stimulate the immune system to destroy cancer cells and the possibility of a synergistic effect between immunotherapy and targeted therapy is under investigation [4]. The CheckMate 025 phase III multicentre trial of 821 patients demonstrated that the programmed (T-cell) death-1 (PD-1) inhibitor, nivolumab, which potentiates the native T-cell response has demonstrated favourable an overall survival benefit with a 27% reduction in the risk of death as well as a greater response rate (25% versus 5%) when compared to treatment with everolimus [14].

Our preliminary work in patients treated with multi-kinase inhibitors has shown that different clinical imaging patterns may be observed with treatment [15]. New metastases may appear even if pre-existing lesions remain stable. Changes in size and/or vascularisation of existing lesions may occur either rapidly or slowly. Proposed vascular mechanisms of resistance include: 1) Up-regulation of alternative pro-angiogenic pathways, allowing the development of new abnormal tumour vessels, restoring tumour angiogenesis to its pre-treatment level and allowing rapid tumour growth; 2) Development of alternative metabolic pathways independent of vascular supply, which may represent a more aggressive phenotype; 3) Recruitment of vascular progenitor cells, pro-angiogenic monocytes, pericytes, or co-option of normal vessels, thus reconstituting vessels which do not have the typical characteristics of tumour vessels. These pseudo-normal vessels are inefficient resulting in a slower tumour re-growth rate.

Assessing response to targeted therapies with the internationally accepted response evaluation criteria in solid tumours version 1.1 (RECIST 1.1) has been challenging in this setting. With RECIST 1.1 percentage changes in the sum of maximal unidirectional measurements of target visceral or short axis nodal metastatic lesions over time are used to categorise treatment response, disease stability or progression [16]. However targeted therapies do not necessarily lead to a concomitant size reduction [15, 17]. Indeed immunotherapies such as ipilimumab and pembrolizumab may also initially increase lesion size, in up to 15% of cases during successful therapy (‘pseudoprogression’) [18]. This may be due to the accumulation of inflammatory infiltrates secondary to the augmented immune response [19]. One such study of Pembrolizumab in 411 patients treated for metastatic melanoma found that 12% of patients demonstrated an initial increase in metastatic lesions that would have led to a classification of progressive disease by RECIST 1.1 despite a decline in overall tumour burden over time [20, 21]. In order to prevent premature cessation of efficacious treatment and conversely continued treatment with ineffective treatment there is a need to improve imaging response assessment. An example of preliminary work is the incorporation of changes in tumour density (Hounsfield unit attenuation) measured on contrast enhanced computed tomography (CT) with size change [22, 23] and immune response criteria [20].

We propose that in vivo imaging targeted at the mode of drug action, its downstream effects, and a better understanding of the mechanisms underlying tumour response and resistance will enable imaging to provide a better indication of therapeutic effect. Hybrid positron emission tomography/magnetic resonance imaging (PET/MRI) scanners enables PET tracers such as the glucose analogue fluorine 18 (¹⁸F) fluorodeoxyglucose (FDG) to be combined with multi-parametric MRI including multiphasic contrast enhanced and diffusion weighted sequences. Preliminary studies of ¹⁸F–FDG PET alone [24‐30] or MRI alone [31‐34] have suggested a role for these modalities as response biomarkers. ¹⁸F–FDG PET has a high sensitivity and specificity for extra-renal lesions with a sensitivity and specificity of up to 84% and 91% in detecting metastases, which appear hypermetabolic and and a Phase II study of 44 patients has shown that reduction in the FDG metabolic activity and tumour size is associated with overall and progression free survival [24]. A high baseline FDG standardised uptake value (SUV) also correlates with disease aggressiveness [26, 35, 36].

The REMAP study will examine whether the performance of integrated FDG-PET/MRI is better than standard of care imaging for response categorisation in metastatic RCC patients undergoing targeted therapy. We hypothesise that integrated ¹⁸F–FDG PET/MRI combining anatomic, metabolic and physiological imaging will be able to detect response/resistance ahead of size change allowing non-responding patients to transition to alternative therapies more swiftly. An integrated whole body technique will also allow for the heterogeneity of response to be explored within and between lesions, building on existing work [37]. PET/MRI will also facilitate the assessment of potential effects of therapy e.g. angiogenesis inhibition or immune modulation on normal organ microvasculature, which underlie treatment co-morbidities, e.g. hypertension as suggested in our previous work [38].

The REMAP study is an investigator initiated single centre observational study. All participants will have to provide written informed consent, signed and dated before inclusion in the study. The study is registered on the NIHR CRN portfolio ISRCTN12114913.

The sponsor is King’s College London and Guy’s & St Thomas’ Hospitals NHS Trust, London. The study is funded by Cancer Research UK (CRUK award reference C42827/A20000) and will be carried out at the King’s College London and Guy’s and St Thomas’ PET Centre, London.

The final protocol has been approved by the Southeast London Research Ethics Committee (ref: 16/LO/1499) and approved by the Administration of Radioactive Substances Advisory Committee of the Dapratment of Health (ARSAC), UK.

Thirty eight patients with proven metastatic renal cell carcinoma will be recruited into the study. Patient inclusion and exclusion criteria are summarised in Table 1. A staging CT is required. Study entry will be restricted to patients in whom there is at least one lesion 2 cm or greater in size in order to minimise partial volume effects in ¹⁸F–FDG PET/MRI.

The primary objective of the REMAP study is to compare integrated ¹⁸F–FDG PET/MRI with CT for response categorisation at 12 weeks and 24 weeks on treatment and to assess the sensitivity and specificity of integrated ¹⁸F–FDG PET/MRI metrics to predict for disease progression after 12 and 24 weeks on treatment (PET: FDG standardised uptake value (SUL_peak); MRI: diffusion MRI apparent diffusion co-efficient (ADC) and post-gadolinium contrast agent normalised tumour signal intensity (SI_norm), (alone or in combination).

Secondary objectives are; 1) to compare ¹⁸F–FDG PET/MRI response categorisation versus RECIST 1.1 and exploratory combined size & enhancement criteria (e.g. Choi, modified Choi and MASS [39‐41]; 2) to assess if whole body tumour burden is better than target lesions only for response categorisation (RECIST 1.1); 3) to assess and quantify the inter-lesional heterogeneity in responding/non-responding lesions by texture analysis and 4) to explore whether there is a consistent effect of therapy on normal organ physiology e.g. kidney, liver, spleen.

The primary endpoint is the number of responders/non-responders on ¹⁸F–FDG PET/MRI versus standard CT at 12 and 24 weeks and a consensus reference standard based on all patient information up to 36 weeks. Time to progression and progression free survival beyond 36 weeks will also be recorded in order to determine if ¹⁸F–FDG PET/MRI metrics can predict for disease progression.