Metabolic changes in RCC

Renal cell tumours harbour significant metabolic perturbations and different histologic subtypes have distinct metabolic phenotypes. In clear cell RCC (ccRCC), the major genetic driver is the loss of the von Hippel-Lindau tumour suppressor gene, which leads to accumulation of hypoxia-inducible factor 1α with downstream transcriptional activation of pathways involved in glycolysis.^{11 12} Metabolomic analyses have confirmed increased lactate labelling after [U-¹³C]glucose infusion in ccRCC tumours compared with adjacent normal kidney with glycolysis increasing in a grade-dependent manner.^13,15 On the other hand, renal oncocytomas, which are classified as benign renal masses, suppress oxidative metabolism due to defective complex I within the mitochondrial electron transport chain.^1116,18 Therefore, it is not clear to what extent the ratio between glycolytic and oxidative metabolism varies across aggressiveness of renal neoplasms.^{17 19} Recently, studies have also reported mitochondrial respiration defects in both renal oncocytoma and its malignant counterpart, chromophobe RCC (chRCC), with the potential for differentiation based on genomic, transcriptomic and metabolic levels.^{16 17 20} The question to be addressed in this study is whether MRI can be used to phenotype the aggressiveness of renal masses, and can therefore implement to help risk-stratify tumours to inform management decisions.

Imaging metabolism

Building on the evidence described above, we hypothesise that novel MRI techniques can non-invasively characterise whole-tumour metabolism and its heterogeneity across renal tumours. The most common clinical tool for assessment of tumour metabolism uses a radiolabelled glucose analogue, in conjunction with positron emission tomography, but its effectiveness is limited by renal excretion of the tracer, and it has failed to show success in characterising renal tumours.^{21 22 99m}Tc-sestamibi in conjunction with single-photon emission CT has been used to detect a variable degree of tracer uptake in renal oncocytoma compared with chRCC in a small number of patients, which has been attributed to the tracer accumulation in cells with high levels of mitochondria.^{23 24} However, both techniques expose patients to ionising radiation, offer poor soft tissue contrast and the inability to detect specific downstream metabolites or their metabolic compartmentalisation.^{25 26}

Our multidisciplinary group are developing novel non-radioactive and non-invasive clinical tools for imaging tumour metabolism. For example, hyperpolarised [1-¹³C]-pyruvate MRI (HP ¹³C-MRI), following injection of intravenous hyperpolarised ¹³C-pyruvate, can be used to simultaneously detect glycolytic metabolism in the cytosol and oxidative metabolism in the mitochondria in tissue where there is sufficient metabolism.²⁷ Two recent papers have shown that ¹³C-lactate labelling can be used to distinguish high-grade ccRCC from lower-grade tumours, driven by the pyruvate transporter (monocarboxylate transporter 1; MCT1).²⁸ Furthermore, a case of a renal oncocytoma displayed the lowest pyruvate-to-lactate conversion compared with a range of malignant masses, including ccRCCs, suggesting that this may be a tool for discriminating these lesions.²⁸

More recently, we have implemented deuterium (²H) metabolic imaging (DMI) at clinical field strength as an alternative method for non-invasive detection of metabolism using orally administered deuterated glucose.²⁹ This method can also be used to detect cytosolic lactate formation and mitochondrial oxidative metabolism, and is complementary to the information provided by HP ¹³C-MRI.^{30 31} We have undertaken this in the brain at clinical field strength (3T) and will apply it to the kidney in this study using dedicated hardware to assess whether differential glucose metabolism can be used to differentiate benign from malignant lesions.

Imaging cellularity and structure

To complement the metabolic measurements, we will evaluate non-invasive methods to distinguish benign from malignant lesions based on differences in cellularity on histology. Sodium MRI (²³Na-MRI) is a complementary tool to probe tissue structure as a measure of the tissue sodium concentration, which is a function of cellularity due to the concentration gradient across the cell membrane. The method can also be used to extract an intracellular-weighted component of the sodium pool as a measure of the transmembrane sodium gradient.³² In the presence of hypoxia, the sodium/potassium ATP pump may be inhibited, leading to alterations in the sodium concentration.^{32 33} We have previously shown how ²³Na-MRI can be used to demonstrate the Na⁺ gradient across the corticomedullary axis in the normal kidney and how this can be used to assess dynamic changes in renal sodium.³³ We have also recently developed a novel birdcage MRI coil system for high resolution ²³Na-MRI of the normal kidneys,³⁴ which we will apply to imaging focal renal pathology as part of this study. We have shown the potential of ²³Na-MRI in several cancer types, correlating with cellularity on histology, and will apply this to small renal masses for the first time here.^{35 36}

In addition, we have developed complementary ¹H-MRI methods to quantitatively map T2 relaxation properties within tissue for assessment of diffuse renal disease.³⁷ Here, we will assess whether these can distinguish benign from malignant lesions. These parameters are dependent on the local tissue chemical properties and reflect the microenvironmental differences between benign and malignant disease.

Study design and objectives

The investigation of differential biology of benign and malignant renal masses using advanced MRI techniques study is designed as a feasibility study to acquire preliminary data with set imaging protocols as defined based on our previous work, and these results will be used to optimise future imaging protocols and to inform large-scale studies. The study will be conducted as a non-randomised, physiological imaging study in patients with localised renal masses, at a single site: Addenbrooke’s Hospital, Cambridge University Hospitals NHS Foundation Trust. For all imaging studies, the primary objective will be the technical development of each imaging technique with ultimate aim to understand whether the methods can identify any differences between benign and malignant tumours, including measuring the ¹³C-pyruvate-to-lactate conversion with HP ¹³C-MRI, quantifying sodium concentration using ²³Na-MRI, detecting ²H-glucose and its metabolites using DMI. Secondary objectives will aim to compare the imaging techniques to understand if they can produce complementary information. As an exploratory part of the study, the objective will be to link the imaging data with clinical data and tissue molecular analyses for biological validation of the novel MRI techniques. Diagnostic accuracy analysis is not foreseen at this stage.

Participant selection

Participants will be identified through multidisciplinary team meetings or by clinical teams involved in their routine care at Addenbrooke’s Hospital, and recruited if they meet all the inclusion and none of the exclusion criteria as detailed in table 1. The participants will be allocated into three imaging arms with 10 patients in each: (1) HP ¹³C-MRI; (2) DMI; and (3) ²³Na-MRI. ¹H-MRI with T2-mapping will be performed in all patients. The recruited participant will be allocated to one of these imaging techniques by the chief investigator. This will be based on the availability of kits for HP ¹³C-MRI, availability of the deuterated glucose drink for the DMI study and/or availability of the research MRI scanner. The aim is to recruit at least four patients in each arm with an oncocytic renal neoplasm (mostly oncocytomas) and at least four patients with RCCs as determined by clinical pathology assessment to enable direct comparison. Half of the patients in each arm will be selected from a cohort of newly diagnosed renal masses and imaged prior to biopsy, with 75%–80% of these expected to have RCCs, and most of the remainder having oncocytic neoplasms. The other half will be acquired from retrospective cohorts of patients with previously diagnosed oncocytic neoplasms on active surveillance at least 6 weeks post biopsy. Diagnosis will be made on tissue samples acquired at biopsy or at surgery if applicable, using molecular markers where possible. This approach will ensure an appropriate balance between benign and malignant lesions in each cohort.

Participants may be removed from the study at their choice or at the investigator’s discretion if it is felt to be clinically appropriate. Reasons for participant withdrawal will be recorded. Primary reasons for withdrawal may include serious adverse event (SAE), withdrawal of consent, lost to follow-up, participant non-compliance, or study closed or terminated. Participants who are withdrawn from the study or do not complete at least one scan will be replaced.

Interventions

Study participants will be deemed evaluable if they receive at least one scan on any of the three imaging techniques. Each study participant will be allocated a unique study number following study enrolment and will be identified by this number throughout the data collection and analysis process.

The participants will be asked to attend all or some of these timepoints:

1. Baseline imaging visit.

2. An optional repeat scan within 7days of the first scan using the same imaging technique.

3. For those not taking part in part 2 above, an optional scan with another imaging technique within 14 days of the first scan. The second imaging arm will be determined as detailed above; that is, by the chief investigator based on the availability of the imaging kits and the research MRI scanner.

4. An optional research biopsy will be undertaken at standard of care surgery or at biopsy, whichever will be clinically indicated and performed.

The study flow chart is presented in figure 1.

Open in a separate window

Figure 1

Study flow chart. DMI, deuterium metabolic imaging; HP ¹³C-MRI, hyperpolarised [1-¹³C]-pyruvate MRI; MRSI, magnetic resonance spectroscopy imaging; ²³Na-MRI, sodium MRI.

Table 2 provides an overview of assessments to be performed at each study visit.

Table 2

Schedule of assessments

Assessment	Visit/days
	Screening†	Baseline MRI visit	Optional repeat MRI visit within 7days of baseline (if applicable)	Optional MRI visit with different imaging technique within 14 days of baseline (if applicable)
Attend unit	*	*	*	*
Consent‡	*	*	*	*
Medical history	*
Demography (weight, height, sex, date of birth)	*
Clinical examination	*
ECOG performance score	*	* if clinically indicated	* if clinically indicated	* if clinically indicated
Venous blood sample§¶	*	*	*	*
Pregnancy test in WOCBP	*	*	*	*
MRI scan		*	*	*
General/additional assessments		* if clinically indicated	* if clinically indicated	* if clinically indicated
Vital signs
Pre imaging		*	*	*
Post imaging		*	*	*
Injection of 13C-pyruvate and/or deuterated glucose drink (for 13C-pyruvate and DMI techniques only)		*	*	*

Open in a separate window

^*Applicable assessment.

^†We will attempt to screen for participants during their standard of care visits and use of medical records.

^‡Ongoing consent will be confirmed at each visit.

^§Venous blood samples will include, but may not be limited to; biochemical series and liver function test.

^¶Full blood count samples will only be required at injection timepoints if they have not been taken recently, that is, within 14 days. These results will be copied in the study data.

DMIdeuterium metabolic imagingECOGEastern Cooperative Oncology GroupWOCBPwoman of childbearing potential

For participants with benign renal masses that subsequently undergo active surveillance and have repeated MRI scans, we will seek the permission from the participants to review the clinically required MRI scan and compare to what was collected at the research scan. If the participant is not due to have a clinically required MRI scan, the research team will not affect this decision.

We will endeavour to minimise the number of injections and/or deuterated drinks for each participant, in as, the maximum number of injections/drinks will be limited to two each. Routes of administration for each of the imaging agents are as following:

1. HP ¹³C-MRI: single intravenous injection of up to 40mL at 0.4mL/kg of ¹³C-pyruvate through an intravenous cannula while the study participant is in position on the scanner bed. Scanning will begin 12s after the injection. There will be an optional repeat injection of ¹³C-pyruvate at all imaging visits to test for repeatability of the ¹³C-MRI technique. These will take place within 7days of the first scan.

2. ²³Na-MRI: not applicable. No additional research probe will be given to the participant as part of this imaging technique.

3. DMI: at each imaging visit, the participant will receive a deuterated glucose solution where 60g of glucose is dissolved in 200mL of water for injection and the dose solution will be adjusted to their body weight at 0.75g/kg body weight. Scanning will begin 60min after the drink ingestion. There will be an optional repeat deuterated drink at all imaging visits to test for repeatability of the DMI technique. These will take place within 7days of the first scan.

The sample size calculation and outcomes

The study has been powered to assess changes in the ¹³C-pyruvate metabolism from the data we collected from nine treatment-naïve renal tumour patients.²⁸ This work showed that the median pyruvate-to-lactate conversion constant (k_PL) in ccRCCs was 0.0065 (range 0.0024–0.0151), while it was 0.0043 (range 0.0028–0.0076) in the normal kidney. This study also reported metabolism in a single case of renal oncocytoma, which showed both the lowest conversion constant and lactate-to-pyruvate ratio (LAC/PYR). There are currently no published studies assessing DMI and ²³Na-MRI quantitative parameters in human kidney tumours.

Based on the parameters obtained from the HP-¹³C-MRI, pragmatic sample sizes have been determined for priming of the study: we plan to include up to 30 participants in total: 15 with benign renal masses and 15 with malignant renal masses. These participants will be divided equally into three imaging arms (HP ¹³C-MRI, ²³Na-MRI and DMI); therefore five benign and five malignant participants will be recruited to each imaging arm. If participants are willing to take part in the optional additional scan using a different imaging technique, these participants will be counted towards both arms of the study and therefore the total number recruited to the study will be less than 30 participants. The planned timeline for the study is as follows: start date on 1 January 2023, primary completion date on 31 August 2025, with study completion by 1 January 2026.

Descriptive statistics will be used. The primary covariates to be studied are as follows:

1. HP-¹³C-MRI: ratio of the summed hyperpolarised ¹³C-lactate to the summed ¹³C-pyruvate over the timecourse of the experiment as a quantitative metric of pyruvate-to-lactate exchange catalysed by the enzyme lactate dehydrogenase. This metric is termed the LAC/PYR. We have significant experience in developing quantitative methodology to analyse this data.⁴⁰

2. ²³Na-MRI: total sodium concentration, as a metric to quantify accumulation of Na+inthe tissue of interest. This metric was used in comparison between prostate cancer and normal prostate tissue.³⁵

3. DMI: primary goal is the technical development of the abdominal DMI, which has not been extensively developed yet due to limitations in detection of metabolites within the DMI spectrum in abdomen attributed to lipid peaks and variability of tissues. However, we aim to evaluate the ratio of the summed ²H-lactate over the summed combined signal from ²H-glutamine+²H-glutamate as a measure of the ratio of glycolysis to oxidative metabolism, as previously shown in healthy human brain.⁴¹

Data management and confidentiality

Ethics and dissemination

Safety considerations

Patient and public involvement (PPI)

Patient representative groups were closely involved in preparation of the study protocol. We have sought help from the Cambridge University Hospitals PPI Panel, who kindly reviewed the study documentation we were intending to submit with the IRAS for the REC review. With their valuable feedback, we have adapted documentation to make it easier to follow, such as developing graphic representations of the study procedures and preparing separate documents for each of the imaging arms.

We acknowledge the invaluable feedback by the patient representatives, as well as the administrative and technical support from the Advanced Cancer Imaging and Urological Malignancies programmes, Cancer Research UK Cambridge Centre, and radiographers of the Magnetic Resonance Spectroscopy Unit, Addenbrookes.