Multiparametric MRI for assessment of early response to neoadjuvant sunitinib in renal cell carcinoma.

PURPOSE: To detect early response to sunitinib treatment in metastatic clear cell renal cancer (mRCC) using multiparametric MRI.
METHOD: Participants with mRCC undergoing pre-surgical sunitinib therapy in the prospective NeoSun clinical trial (EudraCtNo: 2005-004502-82) were imaged before starting treatment, and after 12 days of sunitinib therapy using morphological MRI sequences, advanced diffusion-weighted imaging, measurements of R2* (related to hypoxia) and dynamic contrast-enhanced imaging. Following nephrectomy, participants continued treatment and were followed-up with contrast-enhanced CT. Changes in imaging parameters before and after sunitinib were assessed with the non-parametric Wilcoxon signed-rank test and the log-rank test was used to assess effects on survival.
RESULTS: 12 participants fulfilled the inclusion criteria. After 12 days, the solid and necrotic tumor volumes decreased by 28% and 17%, respectively (p = 0.04). However, tumor-volume reduction did not correlate with progression-free or overall survival (PFS/OS). Sunitinib therapy resulted in a reduction in median solid tumor diffusivity D from 1298x10-6 to 1200x10-6mm2/s (p = 0.03); a larger decrease was associated with a better RECIST response (p = 0.02) and longer PFS (p = 0.03) on the log-rank test. An increase in R2* from 19 to 28s-1 (p = 0.001) was observed, paralleled by a decrease in Ktrans from 0.415 to 0.305min-1 (p = 0.01) and a decrease in perfusion fraction from 0.34 to 0.19 (p<0.001).
CONCLUSIONS: Physiological imaging confirmed efficacy of the anti-angiogenic agent 12 days after initiating therapy and demonstrated response to treatment. The change in diffusivity shortly after starting pre-surgical sunitinib correlated to PFS in mRCC undergoing nephrectomy, however, no parameter predicted OS. TRIAL REGISTRATION: EudraCtNo: 2005-004502-82.

Introduction

Renal Cell Carcinoma (RCC) is the most common malignant tumor of the kidney, with prognosis being highly stage-dependent. Despite the increasing detection of early-stage RCC as an incidental finding on cross-sectional imaging, 15–20% of patients still present with advanced, stage IV disease [1, 2]. 5-year overall survival (OS) exceeds 80% in patients diagnosed with stage I and II RCC, however, the prognosis of patients with locally advanced or metastatic RCC (mRCC) has a far lower 5-year OS at 11.3% [1, 2].

The therapeutic spectrum for advanced and metastatic RCC has expanded rapidly over the past decade. Whilst vascular endothelial growth factor (VEGF)-targeted therapy combined with an immune-checkpoint inhibitor has become the standard-of-care in fit patients with mRCC, (VEGF)-targeted therapy like sunitinib continues to be used in all lines of treatment [3, 4]. Expansion in the therapeutic options of mRCC requires a parallel improvement in methods of early detection of therapy response to optimize the time spent on effective therapy and reduce side-effects and costs in the absence of clinical benefit.

Current guidelines suggest two to four-monthly follow-up of mRCC patients on systemic therapy with contrast-enhanced CT for response assessment according to the Response Evaluation Criteria in Solid Tumors (RECIST 1.1) [5]. However, these criteria have well-known limitations, and there is no clinical evidence that RECIST-defined disease progression is a clinically valid endpoint that requires treatment interruption or modification [3]. Additionally, patterns of treatment response to immunotherapy and targeted therapy differ from cytotoxic chemotherapy where volume-effects predominate, and are insufficiently captured by RECIST at early time points [6]. While survival is correlated with volumetric response in mRCC, even stabilization of the disease is a valuable achievement, which is, however, difficult to detect morphologically [7, 8].

An early marker of response to VEGF-targeted treatment with tyrosine kinase inhibitors (TKI) would allow a switch to alternative anti-cancer therapy before the tumor burden increases in patients who derive no benefit from TKI. The mechanism of action of VEGF-targeted TKI like sunitinib, namely inhibition of tumor angiogenesis and vascular maintenance, supports the use of physiological MRI for the early detection of treatment response [9]. Intravoxel incoherent motion-type diffusion-weighted imaging (IVIM-DWI) can differentiate the contribution of microperfusion and tissue diffusion to the overall diffusion signal [10]. Meanwhile, blood oxygen level-dependent (BOLD) MRI is sensitive to tissue hypoxia, which has been shown to increase following sunitinib treatment in melanoma xenografts [11, 12]. Finally, tissue perfusion can be assessed using dynamic contrast-enhanced (DCE) MRI.

The safety and efficacy of 12 days of pre-surgical sunitinib were investigated in the NeoSun prospective clinical trial (EudraCT No: 2005-004502-82). Sunitinib was chosen as a well-investigated first-line agent in mRCC and for its potential as a neoadjuvant agent to reduce the volume of the primary tumor [13]. This report investigates the effects of pre-surgical sunitinib therapy using quantitative, physiological and morphological MRI measurements and assesses whether these can be used to detect early treatment response and predict overall and progression-free survival in the NeoSun cohort.

Materials and methods

Participant selection

Imaging data from the NeoSun trial, a prospective, single-centre, single-arm phase II study of neoadjuvant sunitinib in mRCC, was assessed following a pre-planned analysis [14]. National research ethics committee approval was obtained (REC ref: 2005-004502-82) and written informed consent was provided by all study participants. Key inclusion criteria were histopathological confirmation of clear cell RCC with metastases judged by the treating clinician to potentially derive benefit from sunitinib prior to a planned cytoreductive nephrectomy. Participants were excluded if they had undergone previous treatment for mRCC or were unable to undergo MRI. Twenty-two consecutive participants were screened for inclusion in the trial between 04/2011 and 01/2014. Six participants were excluded prior to enrolment and four participants did not fulfil the criteria for study completion (see CONSORT diagram, Fig 1). This manuscript reports the outcomes of the exploratory imaging endpoints of the NeoSun trial. Primary oncological outcomes of the trial have been published elsewhere [14] and only participants who completed the full MR-imaging protocol are included in this report. One participant did not tolerate the MRI. The incomplete dataset was of poor quality and, therefore, excluded from further analysis.

Fig 1

CONSORT diagram explaining recruitment and reasons for exclusion of participants.

Treatment

All participants underwent treatment with 50 mg sunitinib once daily for 12 days prior to radical cytoreductive nephrectomy; one participant due additionally underwent bilateral adrenalectomy for metastatic involvement. Sunitinib therapy was resumed after surgery on a repeating 6-week cycle (4 weeks-on, 2 weeks-off) until RECIST 1.1 defined disease progression was identified on 12-weekly follow-up CT, or patients experienced unacceptable toxicity, as assessed by the National Cancer Institute Common Terminology Criteria for Adverse Events, Version 4.0 [15].

Survival

Overall survival (OS) was defined as the time between enrolment and the date of death from any cause. Participants who were lost to follow-up were censored at the last date when they were known to have been alive. Patients who were still alive on the 01/03/2020 were censored on this date, resulting in a median follow-up of 2.6 years (range: 0.6–8 years) for all patients and between 2.7 years and 8 years for surviving patients. Progression-free survival (PFS) was defined as the time between enrolment and the date of RECIST 1.1 defined progressive disease, or death from any cause.

MRI technique

Multiparametric MRI was acquired at baseline and after 12 days of pre-surgical sunitinib treatment. Treatment response assessment was performed using contrast-enhanced CT of the thorax, abdomen and pelvis at baseline and every three months following nephrectomy, until disease progression.

MR imaging was performed on a 1.5 T Discovery MR450 system (GE Healthcare) using an 8-channel cardiac array coil. The multiparametric protocol included sagittal DWI using a dual-spin-echo echo-planar imaging sequence and b-values of 0, 150, 500, 700 and 900 s/mm2 with TE 73.4 ms; TR 4000 ms; FoV 35x35 cm2; slice thickness/gap 4/1 mm; acquisition matrix 104x104; 6 Nex; receiver bandwidth ±250 kHz; ASSET factor 2; scan time 5 minutes 16 seconds (Table 1). Trace-weighted images were acquired by averaging 3 orthogonal diffusion directions for the non-zero b-values. Saturation bands were used to reduce signals from outside the volume of interest. Although no specific DWI performance characterisations were performed for this specific study, the reliability of the body DWI measurements was carefully characterised as part of another study conducted on the same MRI system and during an overlapping time-period [16]. The S1 File provides detailed acquisition information for the remaining sequences. The median acquisition time was 61 minutes (table time, range 56–68 minutes). For the DCE-MRI acquisition, 0.1 mmol/kg of Gd-DOTA (Dotrarem, Guerbet) was used.

Table 1

Imaging parameters.

Sequence	TE / TR [ms]	Flip Angle [deg]	Matrix Size	Slice Thickness / Orientation	Other parameters
T2w FRFSE	48–69 / 4000	90	320 x 224	4 mm coronal	ETL: 10–13, RT
				1 mm gap	2 Nex
T1w FSPGR	4.8 / 139	70	256 x 256	4 mm coronal	BH
				1 mm gap	0.75 Nex, ASSET 2
DW-EPI	73.4/4000	90	104 x 104	4 mm sagittal	b-values: 0, 150, 500, 700 and 900 s/mm2
(IVIM type)				1 mm gap
					6 Nex, ASSET 2
R2* mapping	4.76–47.6 (echo spacing 4.76) / 100	25	128 x 128	4 mm sagittal	BH
				1 mm gap	ASSET 2
T1 mapping	1.6 / 3.9	1, 3, 5, 10, 15 and 20	160 x 160	5 mm coronal-oblique	BH
					1 Nex
DCE-MRI	1.6 / 3.9	18	160 x 160	5 mm coronal-oblique	intermittent bh, temporal resolution: 4.3–6.4 s
					0.5 Nex

TE: Echo Time, TR: Repetition Time, ETL: Echo Train Length, RT: respiratory triggered, BH: breath hold, DCE: dynamic contrast-enhanced, Nex: Number of excitations.

Image processing

Analysis of the IVIM DWI data was performed using in-house developed software developed in MATLAB (The Mathworks, version R2010b). The four non-zero b-values were used for voxelwise mono-exponential calculations of the diffusivity D. The perfusion fraction f was then estimated at a voxel level from the fractional difference between the measurements at b = 0 s/mm2 and the extrapolated signal from the higher b-values. A repeatability coefficient of 20% was pre-defined as a threshold for the detection of treatment-related changes based on existing literature describing the repeatability of ADC measurements [17-19].

In-house software developed in MATLAB was used to generate R2* and T2* maps from the multi-echo gradient-echo images. The pixel-level data were fitted to a mono-exponential decay using the nonlinear Levenberg–Marquardt algorithm, and using a log‐linear approximation to compute the initial values for the fits.

The T1 mapping and DCE-MRI data were processed in MIStar (Apollo Medical Imaging Technology). Each dataset was pre-processed using a 3x3 median filter. The T1 mapping and DCE datasets were co-registered both within and between the datasets to remove (as far as possible) spatial mis-registrations induced by motion. The Tofts model [20] using a population-averaged Arterial Input Function [21] was applied to calculate maps of the transfer constant Ktrans and quantify the contrast-concentration-versus-time curve for 90 seconds after the arrival of the contrast bolus (iAUC90).

Regions-of-interest (ROI) were outlined for the normal-appearing ipsilateral kidney, the tumor and its necrotic/cystic parts by two radiologists with 8 (T.B.) and 5 (F.Z.) years of experience on the coronal T2w FRFSE sequence, the sagittal ADC map, the sagittal R2* and coronal-oblique Ktrans maps using ImageSetViewer Software, version 1.7 (University Health Network Toronto). The segmentation was undertaken blinded to the clinical outcome and RECIST measurements. The region of interest for the solid tumor components was generated from the subtraction of the necrotic/cystic parts from the entire tumor. Using custom-written MATLAB software, slice-wise ROIs were combined into whole volumes of interest and used to calculate the volumes of each tumor component (from T2w imaging) and also the median values of Ktrans, iAUC90, R2*, fp and D within the whole and solid tumor components for each participant. The test re-test variation in imaging parameters was compared in the normal renal tissue to appraise the repeatability of quantitative imaging measurements in tissues not targeted by sunitinib. A single radiologist (T.B.) performed the RECIST 1.1 assessment on the CT scans acquired at baseline and following nephrectomy. The contrast-enhanced CT scans covererd the chest, abdomen and pelvis and included metastatic lesions which were not included in the field-of-view of the MRI scan which was restricted to the tumour-bearing kidney.

Statistical methods

Statistical analysis was performed in the SAS suite (version 9.4). Differences before and after therapy were assessed with the non-parametric Wilcoxon signed-rank test. Differences in survival were visualised using Kaplan-Meier plots and survival distributions compared using log-rank tests. The Mann-Whitney U test was used for the comparison of unpaired samples. P values ≤ 0.05 were considered as statistically significant.

Results

The data of thirteen of the 22 participants screened for participation in this study was analyzed. Twelve patients (1 female, 11 male) were included in this study (Fig 1). The median age at enrolment was 60 years (range: 50–74, mean 61 years). The stage at nephrectomy was pT1b in one patient, pT3a in 10 patients and pT3b in one patient. Five patients were diagnosed with pN1 disease and all patients had cM1 disease. Participant characteristics are summarized in Table 2.

Table 2

Participant characteristics.

Characteristic	n = 12
Age at enrolment (mean ± SD)	61 ± 7.6 years
Gender (f/m)	1 / 11
Stage at nephrectomy
pT1a	0
pT1b	1
pT2a	0
pT2b	0
pT3a	10
pT3b	1
pT4	0
Location of metastasis at diagnosis
Bone	1
Lung	10
Pleura	2
Adrenal	2
Lymph nodes	5
Pancreas	1
Overall Survival (median [range])	138 weeks [30–417]
Progression-free survival (median [range])	67 weeks [17–417]

SD: Standard deviation.

Volumetric findings

At baseline, the mean volume of the primary tumor was 633 ml (SD ± 386 ml, range 175–1369 ml) of which 68% (15–95%) equivalent to 388 ml (± 249 ml, 104–903 ml,) was solid tumor tissue. Following pre-surgical treatment with sunitinib for 12 days, an average reduction of 25% to 507 ml (± 351 ml, 90–1213 ml) was observed in the total tumor volume (p < 0.001, df = 11, V = 78, previously reported [14]). Concurrently, the average volume of solid tissue decreased by 28% to 295 ml (± 217 ml, 81–749 ml), p < 0.001 (df = 11, V = 78), with the volume of necrotic/cystic tumor components only reducing by 17%, from 245 ml (± 263 ml, 10–885 ml) to 212 ml (± 251 ml, 8–830 ml), p = 0.005 (df = 11, V = 74). Furthermore, tumor necrosis was associated with decreased response to pre-surgical therapy with a negative correlation between the percentage of necrotic tumor volume at baseline and the overall reduction in tumor volume after 12 days of sunitinib therapy (Spearman rank correlation coefficient ρ = -0.69, 95%-CI: 0.10–0.97, p = 0.02). However, neither the reduction in whole tumor volume nor a solid tumor volume was associated with PFS and OS (Fig 2a/2b). The reduction in volume of the tumor was not associated with the overall RECIST response of the residual metastatic tumor burden following surgery (Fig 3A and 3B). There was a trend for objective response according to RECIST (partial or complete response as best RECIST response) being associated with longer PFS (p = 0.064, median PFS responders: 77 weeks, non-responders: 41 weeks, Fig 2C)

Fig 2

Kaplan Meier survival curves showing no overall survival benefit for participants with larger than median relative tumour volume reduction following 12 days of pre-surgical sunitinib therapy, however, there was a trend for participants with less volume reduction to survive longer (a). No difference in PFS was observed (b) Participants with a larger reduction in tumour diffusivity D experienced longer PFS (c) while there was a trend for participants with objective response according to RECIST 1.1 (partial or complete response) to show longer PFS than patients without objective response (stable or progressive disease) (d). P-values for log-rank test.

Fig 3

Treatment related changes: Percentage change in the whole tumour volume (a), solid tumour volume (b), solid tumour Diffusivity D (c), solid tumour K (d), solid tumour R2* (e) and solid tumour perfusion fraction (f) between patients with stable/progressive disease (SD/PD) and partial/complete response (PR/CR). P-values for Mann-Whitney U test. Relative changes in normal renal tissue illustrate the test re-test variability in imaging measurements.

Diffusion-weighted imaging

Following pre-surgical treatment, a reduction of the median diffusivity D of the solid tumor by 7.7% from 1298 x 10−6 mm2/s to 1200 x 10−6 mm2/s was observed (p = 0.02, df = 11, V = 68). Only two patients achieved a greater than 20% change in D, which was the a priori defined threshold for treatment effect in an individual patient. However, a greater than median reduction in solid tumor D was associated with prolonged PFS (p = 0.031), median PFS 183 weeks vs. 48 weeks, Fig 2D). The baseline D was not predictive of OS or PFS and the reduction in D was not predictive of OS. Concurrently, the perfusion fraction of the solid tumor decreased by 19% from 0.24 to 0.19 (p < 0.001), however, a reduction in perfusion fraction exceeding the predetermined cut-off of 20% was not predictive of prolonged OS (p = 0.88).

The change in mean tumor diffusivity D between baseline and follow-up showed a differential response between participants, with an 11% reduction in patients demonstrating an objective response according to RECIST (partial or complete response; n = 9), and a 3% increase for non-responders (n = 3);. Fig 3C The perfusion fraction fp was not able to differentiate between responders and non-responders and neither D nor fp was associated with long-term treatment benefits.

Blood oxygenation dependent imaging

The median R2* from BOLD MRI increased within the solid parts of the tumor from 19 s-1 to 28 s-1 (p = 0.001, df = 11). However, an R2* at baseline above the median did not predict OS and PFS and the increase in R2* was not associated with improved response according to RECIST.

T1 map

A post-hoc analysis showed a median T1 of 1238 ms (interquartile range: 1184–1341) in the solid tumour at baseline which reduced to 1102 ms (994–1221 ms) after 12 days of sunitinib. The median reduction by 6% was not significant (p = 0.094). Neither the tumour T1 at baseline nor its change following treatment was associated with the RECIST response or survival.

Dynamic contrast-enhanced imaging

Median Ktrans and iAUC90 both showed significant decreases within solid tumor components following treatment initiation, with a mean reduction in Ktrans from 0.415 min-1 to 0.305 min-1 (28% reduction, previously reported [14]), and of iAUC90 from 0.55 to 0.36 mM min (38% reduction), respectively (p = 0.01 and 0.003, df = 11, V = 71 and 78).

Nine participants had a reduction in Ktrans greater than the pre-determined repeatability coefficient of 20% which was selected based on existing literature [22, 23] and is in agreement with more recent findings [24]. However, these participants did not derive an overall or PFS benefit from this.

A typical example of the evolution of all physiological imaging parameters during the pre-surgical sunitinib treatment can be seen in Fig 4.

Fig 4

Examples of the multiparametric MRI before and after 12 days of sunitinib therapy in a 52 year-old male patient.

The parametric maps before and after therapy are windowed equally. This patient showed an overall tumour volume reduction of 14%, with a decrease in volume of the solid tumour fraction by 33% and an increase in the necrotic/cystic tumour fraction of 8% after two weeks of therapy. Additionally, there was a reduction in diffusivity and perfusion fraction as well as an increase in R2* in the solid tumour component. Ktrans decreased post-therapy, which is consistent with the mechanism of action of sunitinib and representative of most of the study patients. D map (diffusivity) and Perfusion fraction from IVIM-type DWI, Ktrans: Contrast transfer constant, R2*: effective transverse relaxation rate, T2w: T2-weighted image.

Discussion

This study showed that early treatment response to pre-surgical sunitinib in RCC can be detected after 12 days using multiparametric MRI and that changes in diffusivity are associated with the volumetric response to treatment and progression-free survival. The data for this study was gathered in a prospective clinical trial where treatment and imaging were tightly controlled and the consistent timing of the follow-up MRI relative to the start of treatment was critical for assessing the early treatment response. Furthermore, the physiological parameters investigated in this study were functionally linked to the mechanism of action of sunitinib and could, therefore, report on the engagement of the drug with its target. The imaging data were acquired on a single MRI-scanner, eliminating inter-scanner variability as a confounding factor.

The reduction in tumor volume following sunitinib therapy was greatest in the solid tumor fraction. This may aid treatment planning with initial neo-adjuvant systemic therapy acting to debulk tumor volume in patients with large tumors, to reduce surgical challenges and risks. In these patients, the presence of large solid tumor components would suggest that a chemoreduction prior to cytoreductive surgery may be feasible with a short course of TKI therapy. Furthermore, the decrease in solid tumor volume could serve as a sensitive biomarker of treatment response. However, only the reduction in diffusivity was associated with PFS, with significant reductions in excess of 20% from the baseline values being observed in two patients. This result will need validation in future, larger cohorts and if confirmed, this imaging approach might be particularly relevant in localized disease where surgery is more impactful as a cure.

The changes observed in physiological imaging parameters were compatible with the expected mechanism of action of sunitinib. Consequently, these widely available MRI techniques could be useful surrogate markers for confirming successful engagement of anti-angiogenic agents with their biological target. Consistently, the perfusion fraction on IVIM-type DWI was reduced, representing a decrease in capillary volume; this was, however, not correlated with survival. This is expected given the decreased activation of the vascular endothelial growth factor receptors (VEGFR1 –VEGFR3) which leads to reduced angiogenesis and vascular maintenance [25]. Furthermore, the R2* was increased, representing decreased oxygenation and vascularization, and the DCE transfer constant Ktrans and area under the contrast-enhancement curve (iAUC90) were decreased, signifying decreased perfusion and vascular permeability following treatment. These imaging surrogates were in agreement with the decrease in immunohistochemical microvessel density observed in the same patient cohort [14].

Previous attempts at predicting treatment outcome in patients undergoing 15 days of sunitinib treatment for mRCC have shown an increased time-to-peak using contrast-enhanced ultrasound (CEUS). This was correlated with the RECIST response after 12 weeks, PFS and OS in a cohort of 38 patients [26]. However, follow-up results including a multi-center study involving 157 patients were not able to confirm these initial results [27, 28]. Contrary to these studies, sunitinib was the first-line treatment in all patients in NeoSun. MRI, as opposed to CEUS, has several advantages. It is operator-independent and the larger field-of-view allows imaging of the entire tumor volume and potentially multiple metastatic deposits in addition. This is particularly advantageous in mRCC where the response to TKI is known to show intra-patient heterogeneity [29]. Finally, CEUS is reliant on having a good acoustic window and little respiratory motion of the target lesion during the three-minute acquisition.

A larger post-treatment reduction in diffusivity of the solid component of the primary tumor was associated with a better RECIST response in the metastatic burden and longer PFS. The overall diffusion restriction increased, despite increasing histological necrosis which acts to increase ADC values [14]. Other studies have demonstrated increases in ADC values after three days of treatment which returned to baseline after ten days and remained unaltered after 3–4.5 months [30, 31]. In the current study, diffusivity was assessed in solid tumor components, while the others analyzed the entire tumor. Given the very heterogeneous nature of RCC, measuring the true diffusivity change following treatment is difficult. The assessment of the solid components alone is likely to be the most accurate measure of the response to therapy in the viable tumor. Furthermore, IVIM-type DWI provided information on diffusion restriction and relative capillary volume simultaneously, whereas previous studies investigated ADC, an aggregate of tissue diffusion and microperfusion [30, 31]. It is conceivable that these parameters undergo sequential changes as the tissue becomes edematous, necrotic, and is then restructured [31]. Furthremore, sunitinib is associated with increasing infiltration of tumour with CD8+ T-cells, which may also lead to a reduction in diffusivity [32]. However, repeat diffusion-weight imaging at more frequent intervals would be needed to resolve these temporal changes.

This study has several limitations, including the small number of patients, however, this is similar to previous literature investigating multiparametric MRI in patients with RCC [33, 34]. The study was stopped ahead of its recruitment target due to slow participant accrual, consequently, the statistical power was decreased. As a single center trial, this will require independent validation to determine generalizability. Furthermore, recent results have shown that cytoreductive nephrectomy in metastatic RCC needs to be considered carefully and is most appropriate in patients with low-volume metastatic disease [3]. In particular, the correlation of imaging biomarkers with OS, a secondary aim of the NeoSun trial, should be investigated in an appropriately powered trial. Recent data has shown that mono-exponential fitting of D0 should employ b-values > 200 s/mm2 to avoid contributions from tissue perfusion [35]. As the data for NeoSun was acquired prior to this publication, the lowest b-value was 150 s/mm2 and perfusion may have had a minor contribution to the overall signal.

Treatment monitoring on standard-of-care CT alone is challenging as tumor size is a late sign of response. In this study, the physiological parameters Ktrans, R2* and fp, which are sensitive to the antiangiogenic effect of TKIs, were significantly altered during early sunitinib treatment. Together with metabolic imaging and radiomics, they can serve as candidate features in future research. Further research on larger cohorts is required to determine the ability of early treatment response detected by MRI to guide treatment decision making and optimize the time-frame of effective therapy.

In conclusion, we demonstrate measurable changes in MRI features which are consistent with the proposed mechanism of action of multi-targeted receptor tyrosine kinase inhibitors, which were measurable after only 12 days of sunitinib therapy and related to the best response according to RECIST and PFS.

Supporting methods.

(DOCX)

Click here for additional data file.

Supporting results.

(DOCX)

Click here for additional data file.

Data.

(DOCX)

Click here for additional data file.

1 Jul 2021

PONE-D-21-15997

Multiparametric MRI for assessment of early response to neoadjuvant sunitinib in renal cell carcinoma

PLOS ONE

Dear Dr. Barrett,

Thank you for submitting your manuscript to PLOS ONE. After careful consideration, we feel that it has merit but does not fully meet PLOS ONE’s publication criteria as it currently stands. Therefore, we invite you to submit a revised version of the manuscript that addresses the points raised during the review process.

ACADEMIC EDITOR:

This is a manuscript of potential interest. However, the Reviewers have raised numerous/major concerns and hence the paper is not acceptable in its present form for publication. Substantial revision along the lines recommended by the Reviewers may render the work suitable for publication. Because the issues raised by the review and requested revisions are rather extensive, my editorial decision will be deferred until completion of the next cycle of review. Please remember that a point by point response to each comment by the Reviewers is absolutely essential, since this is fair to the Reviewers who have put much effort/time into reviewing your paper, and it will definitely help to speed up the review process.

Please submit your revised manuscript by Aug 15 2021 11:59PM. If you will need more time than this to complete your revisions, please reply to this message or contact the journal office at plosone@plos.org. When you're ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

Please include the following items when submitting your revised manuscript:

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). You should upload this letter as a separate file labeled 'Response to Reviewers'.

A marked-up copy of your manuscript that highlights changes made to the original version. You should upload this as a separate file labeled 'Revised Manuscript with Track Changes'.

An unmarked version of your revised paper without tracked changes. You should upload this as a separate file labeled 'Manuscript'.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter. Guidelines for resubmitting your figure files are available below the reviewer comments at the end of this letter.

If applicable, we recommend that you deposit your laboratory protocols in protocols.io to enhance the reproducibility of your results. Protocols.io assigns your protocol its own identifier (DOI) so that it can be cited independently in the future. For instructions see: http://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols. Additionally, PLOS ONE offers an option for publishing peer-reviewed Lab Protocol articles, which describe protocols hosted on protocols.io. Read more information on sharing protocols at https://plos.org/protocols?utm_medium=editorial-email&utm_source=authorletters&utm_campaign=protocols.

We look forward to receiving your revised manuscript.

Kind regards,

Marco Giannelli

Academic Editor

PLOS ONE

Additional Editor Comments:

I would like to note that MR scanner systems should be adequately characterized to obtain reliable quantitative diffusion-MRI measurements. The Authors should adequately discuss this issue, specifying whether they have performed any quality controls to assess the correct application of diffusion weighting gradients and non-linearity effects, which may be relevant for off-center acquisitions of kidneys as well as of the breast.

Journal Requirements:

When submitting your revision, we need you to address these additional requirements.

1. Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming. The PLOS ONE style templates can be found at

https://journals.plos.org/plosone/s/file?id=wjVg/PLOSOne_formatting_sample_main_body.pdf and

https://journals.plos.org/plosone/s/file?id=ba62/PLOSOne_formatting_sample_title_authors_affiliations.pdf

2. PLOS requires an ORCID iD for the corresponding author in Editorial Manager on papers submitted after December 6th, 2016. Please ensure that you have an ORCID iD and that it is validated in Editorial Manager. To do this, go to ‘Update my Information’ (in the upper left-hand corner of the main menu), and click on the Fetch/Validate link next to the ORCID field. This will take you to the ORCID site and allow you to create a new iD or authenticate a pre-existing iD in Editorial Manager. Please see the following video for instructions on linking an ORCID iD to your Editorial Manager account: https://www.youtube.com/watch?v=_xcclfuvtxQ

3. We noted in your submission details that a portion of your manuscript may have been presented or published elsewhere.

[An abstract was presented at ASCO 2017, where drug safety and the duration of overall and progression-free survival were discussed. The abstract is attached to the submission.

A separate manuscript describing the survival outcome, drug safety and translational outcomes based on tissue analysis is currently under review. The only imaging parameters reported in this manuscript are tumour volume and Ktrans. The current, non-peer-reviewed version of the manuscript is attached to this submission.

In contrast to the above publications, the current submission focuses on the translational imaging analysis.]

Please clarify whether this [conference proceeding or publication] was peer-reviewed and formally published. If this work was previously peer-reviewed and published, in the cover letter please provide the reason that this work does not constitute dual publication and should be included in the current manuscript.

4. Thank you for stating the following in the Financial Disclosure section:

[This study has received funding by Pfizer who provided the study drug Sunitinib and an educational grant for the translational endpoint analysis, reported separately.

The project was supported by Cancer Research UK (CRUK), the CRUK Cambridge Centre, Cambridge Trust, the CRUK & Engineering and Physical Sciences Research Council Cancer Imaging Centre in Cambridge and Manchester, the Mark Foundation for Cancer Research, the National Institute for Health Research (NIHR) Cambridge Biomedical Research Centre, Cambridge Experimental Cancer Medicine Centre, Cambridge University Hospitals National Health Service Foundation Trust and the Cambridge Clinical Trials Unit (CCTU), and Addenbrooke’s Charitable Trust.

Pfizer had a right to review the manuscript prior to publication. None of the funders had a role in the study design, data collection or analysis.].

We note that you received funding from a commercial source: Pfizer

Please provide an amended Competing Interests Statement that explicitly states this commercial funder, along with any other relevant declarations relating to employment, consultancy, patents, products in development, marketed products, etc.

Within this Competing Interests Statement, please confirm that this does not alter your adherence to all PLOS ONE policies on sharing data and materials by including the following statement: "This does not alter our adherence to PLOS ONE policies on sharing data and materials.” (as detailed online in our guide for authors http://journals.plos.org/plosone/s/competing-interests).  If there are restrictions on sharing of data and/or materials, please state these. Please note that we cannot proceed with consideration of your article until this information has been declared.

Please include your amended Competing Interests Statement within your cover letter. We will change the online submission form on your behalf.

Please know it is PLOS ONE policy for corresponding authors to declare, on behalf of all authors, all potential competing interests for the purposes of transparency. PLOS defines a competing interest as anything that interferes with, or could reasonably be perceived as interfering with, the full and objective presentation, peer review, editorial decision-making, or publication of research or non-research articles submitted to one of the journals. Competing interests can be financial or non-financial, professional, or personal. Competing interests can arise in relationship to an organization or another person. Please follow this link to our website for more details on competing interests: http://journals.plos.org/plosone/s/competing-interests

5. Please include captions for your Supporting Information files at the end of your manuscript, and update any in-text citations to match accordingly. Please see our Supporting Information guidelines for more information: http://journals.plos.org/plosone/s/supporting-information.

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Partly

Reviewer #2: Yes

**********

2. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: Yes

Reviewer #2: No

**********

3. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: No

Reviewer #2: Yes

**********

4. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

Reviewer #2: Yes

**********

5. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: In this paper authors present a multi-parametric MRI approach to evaluate the response to sunitinib treatment in locally advanced renal cancer.

RECIST criteria, commonly used for evaluating response to therapy are based mainly on tumor size modifications and preset some limitations when immunotherapy or targeted therapy are administrated. Indeed sunitinib in renal cancer inhibits tumor angiogenesis that can early detect with multi-parametric MRI. Nevertheless the interesting and useful approach proposed by the authors, in general there are some issue concerning the technical aspects of this work. Furthermore few patients were included in this study.

In my opinion the study of the impact in terms of OS is not useful in this context. Indeed the primary goal of the study is to find quantitative MRI parameters to evaluate early tumor response. The impact of these on OS could be considered as a further step.

Materials and Methods

Why have you explained in details only DWI protocols while for the other techniques details are provided in table 1?

Lack of information about contrast agent for DCE in Materials and Methods.

What is the total acquisition time?

L142: Please insert Matlab version.

Before IVIM analysis, were images preprocessed for eddy current correction?

Authors state that for diffusion coefficient D calculation all the four non-zero b-bvalues DWI images were considered using a monoexponential fit. Why have you included also the b=150 s/mm2 for D calculation?

Is it well know that for b<200 s/mm2 we have also the contribution of perfusion in diffusion decay (please refers for example to Meeus et al. J. MAGN. RESON. IMAGING 2017;45:1325–1334.) for this reason it was proposed to calculate D including only bvalues higher that 200 s/mm2. Could you explain why have you implemented DWI analysis in this way?

The reference added for population-averaged AIF (ref n. 20) refers to abdomen. Can you argue if this population-averaged AIF can be applied to kidneys? How an altered kidney function could affect AIF?

Statistical methods

L170 Please insert SAS suite version.

Results:

L177: add the reference to fig 1 that provides the diagram with included/excluded patients. It is not clear in this section why 10 patients were excluded.

Reviewer #2: This is a study designed to develop an early imaging biomarker of metastatic RCC response to sunitinib. The use of a homogeneous study cohort, namely subjects with mRCC who all received neoadjuvant sunitinib as first-line therapy, is a significant strength. All scans were acquired on the same scanner, eliminating a potential source of variability. The acquisition of multiple quantitative MRI sequences is another strength, and this was done before and after 12 days of sunitinib treatment. The analysis of this rich image dataset has been done at the ROI/VOI level, rather than the pixel/voxel level, which would have been more powerful. One constraint clearly was that the sequences were acquired in different views (coronal, sagittal, oblique). Overall, this is an interesting study that is potentially useful. Some improvements to the quantitative image analytics would strengthen the report, as suggested below.

Specific Comments:

1) Reproducibility of tumor contouring: What was the Kappa for the two radiologists in defining solid and necrotic/cystic ROIs? An alternate to manual contouring would be to objectively define tumor sub-volumes using the histogram-based approach proposed by Chenevert, T. L. et al. Comparison of voxel-wise and histogram analyses of glioma ADC maps for prediction of early therapeutic change. Tomography 5, 7-14, 2019. https://doi.org/10.18383/j.tom.2018.00049

2) Reproducibility of DCE-MRI: Reference 21 pertains to brain MRI, which may be much more reproducible than MRI of abdominal tumors. And reference 22 pertains to reproducibility of mouse DCE-MRI, which doesn’t seem too relevant to clinical DCE-MRI. I would suggest using the clinical DCE-MRI reproducibility figures published by Galbraith et al. (NMR Biomed 2002 Apr;15(2):132-42. doi: 10.1002/nbm.731.) and Klaassen et al. (Magnetic Resonance Imaging 50:1-9, 2018, https://doi.org/10.1016/j.mri.2018.02.005).

3) Reproducibility of all quantitative MRI: In Figure 3, please also provide comparisons to corresponding changes in MRI parameters pre/post drug in tissues that are presumed to not be affected by drug. This is an option when repeat intra-session scanning was not done as part of the study. See, for example, Lorza, A.M.A., Ravi, H., Philip, R.C. et al. Dose–response assessment by quantitative MRI in a phase 1 clinical study of the anti-cancer vascular disrupting agent crolibulin. Sci Rep 10, 14449 (2020). https://doi.org/10.1038/s41598-020-71246-w.

4) Univariable analysis: Median values of Ktrans, iAUC90, R2*, fp and D within the whole and solid tumor components were analyzed. Was T1 (unenhanced) also evaluated in the same manner?

5) It appears that there were secondary lesions that were large enough for RECIST assessment. What were the post-sunitinib vs. pre-sunitinib quantitative MRI results (analogous to figure 3) from those additional tumors?

6) Discussion: Regarding the observed decrease in ADC of solid tumor following sunitinib treatment, is it known that the short-term response of tumor cells to sunitinib is cell swelling rather than apoptosis?

7) Did the authors perform a multivariable analysis of whether baseline R2*, fp , D, T1 (unenhanced), Ktrans, and iAUC90, taken together (or a subset), are predictive of post- vs. pre-sunitinib change in Ktrans or iAUC90? A change in Tofts model parameters with treatment would be more mechanistically-related to tumor-level drug pharmacodynamics than RECIST, which is a downstream patient-level response metric. Use of delta-DCEMRI rather than RECIST as ground truth for outcome would also allow for the primary plus any metastatic tumors to be analyzed, which would increase sample size. The input data could be augmented to include not just the median values of R2*, fp , D, T1 (unenhanced), Ktrans, and iAUC90, but also additional percentile values (e.g., 10th and 90th) to both improve the predictive performance of the model and to try and do away with the need to manually define “solid” and “necrotic/cystic” tumor regions. Even of the outcome metric remained RECIST, a multivariable analysis would still be useful.

Thank you for the opportunity to review your paper on MRI results from this nice study.

Natarajan Raghunand, PhD

**********

6. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

Reviewer #2: Yes: Natarajan Raghunand, PhD

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email PLOS at figures@plos.org. Please note that Supporting Information files do not need this step.

1 Sep 2021

Reviewer #1: In this paper authors present a multi-parametric MRI approach to evaluate the response to sunitinib treatment in locally advanced renal cancer.

RECIST criteria, commonly used for evaluating response to therapy are based mainly on tumor size modifications and preset some limitations when immunotherapy or targeted therapy are administrated. Indeed sunitinib in renal cancer inhibits tumor angiogenesis that can early detect with multi-parametric MRI. Nevertheless the interesting and useful approach proposed by the authors, in general there are some issue concerning the technical aspects of this work. Furthermore few patients were included in this study.

In my opinion the study of the impact in terms of OS is not useful in this context. Indeed the primary goal of the study is to find quantitative MRI parameters to evaluate early tumor response. The impact of these on OS could be considered as a further step.

Thank you for this comment. The correlation of imaging markers with overall survival was defined as a secondary endpoint in the study protocol. It was included in the protocol since this information is clinically useful. We agree that appropriately powered studies would be required to investigate the relationship between imaging markers and survival further. We have included this as a limitation to our work. (Line 329)

Materials and Methods

1. Why have you explained in details only DWI protocols while for the other techniques details are provided in table 1?

Thank you very much for this question. We have chosen to detail the acquisition parameters in the main manuscript because the parameter which was found to be most useful (D0) is derived from this sequence. However, we have included similarly detailed descriptions of all the other sequences in the supplementary file S1 in addition to table 1. We have included a clarification in the methods section (Line 129).

2. Lack of information about contrast agent for DCE in Materials and Methods.

In this study, 0.1 mmol/kg of Gd-DOTA (Dotarem, Guerbet) was administered during the dynamic series. We have now included this information in the main manuscript in addition to the supplementary information. (Line 131)

3. What is the total acquisition time?

Thank you for raising this important point. The median total acquisition time was 61 minutes and the interquartile range extended from 56 to 68 minutes (n = 24 studies). We have included this information in the manuscript. (Line 130)

4. L142: Please insert Matlab version.

The MATLAB version used to process the data for this manuscript was R2010b, which has now been added to the manuscript. (Line 139)

5. Before IVIM analysis, were images preprocessed for eddy current correction?

Images were acquired using Dual Spin Echo EPI. This minimises eddy current effects so that eddy current correction was not needed.

6. Authors state that for diffusion coefficient D calculation all the four non-zero b-bvalues DWI images were considered using a monoexponential fit. Why have you included also the b=150 s/mm2 for D calculation?

Is it well know that for b<200 s/mm2 we have also the contribution of perfusion in diffusion decay (please refers for example to Meeus et al. J. MAGN. RESON. IMAGING 2017;45:1325–1334.) for this reason it was proposed to calculate D including only bvalues higher that 200 s/mm2. Could you explain why have you implemented DWI analysis in this way?

Thank you for raising this point. We agree that there remains a small perfusion contribution at b = 150 s/mm2 but the effect is much smaller than at lower b-values. This study was planned, and data acquired before the publication of the Meeus paper and therefore the b = 200 s/mm2 was not applied. We chose to neglect the small perfusion contribution at b = 150 s/mm2 and feel that this represents a reasonable compromise for this dataset. We have included this consideration in the limitations of our manuscript. (Line 330)

7. The reference added for population-averaged AIF (ref n. 20) refers to abdomen. Can you argue if this population-averaged AIF can be applied to kidneys? How an altered kidney function could affect AIF?

Thank you very much for this question. The effect of an altered renal function on the applicability of a population-averaged arterial input function will likely depend on the cause of a reduced renal function. A renal artery stenosis would affect both the renal function and the delivery of contrast-agent to the tumour. In contrast, primary or secondary glomerular diseases (e.g. focal segmental glomerulosclerosis, diabetic nephropathy) or obstructive nephropathy would result in a reduction in renal function with a maintained tumour perfusion. The population-averaged AIF described by Parker et al. is used commonly in the quantification of DCE-MRI of RCC 1,2.

Additionally, the primary endpoint of the NeoSun trial was a relative reduction in Ktrans. Therefore, even if the renal function was reduced in a way that had an effect on the contrast-enhancement of the tumour, the renal impairment would be unlikely to progress significantly within the 12-day baseline-to-follow-up period.

Importantly, adequate renal function was an inclusion criterion for the NeoSun trial. Therefore, none of the patients had a serum creatinine concentrations >1.5x the upper limit of the norm.

Statistical methods

8. L170 Please insert SAS suite version.

SAS suite version 9.4 was used for the analyses. This information has been added to the methods section of the manuscript. (Line 171)

Results:

9. L177: add the reference to fig 1 that provides the diagram with included/excluded patients. It is not clear in this section why 10 patients were excluded.

Thank you for this comment, we have added a reference to Figure 1 to line 177 (now line 178).

Reviewer #2: This is a study designed to develop an early imaging biomarker of metastatic RCC response to sunitinib. The use of a homogeneous study cohort, namely subjects with mRCC who all received neoadjuvant sunitinib as first-line therapy, is a significant strength. All scans were acquired on the same scanner, eliminating a potential source of variability. The acquisition of multiple quantitative MRI sequences is another strength, and this was done before and after 12 days of sunitinib treatment. The analysis of this rich image dataset has been done at the ROI/VOI level, rather than the pixel/voxel level, which would have been more powerful. One constraint clearly was that the sequences were acquired in different views (coronal, sagittal, oblique). Overall, this is an interesting study that is potentially useful. Some improvements to the quantitative image analytics would strengthen the report, as suggested below.

Specific Comments:

1) Reproducibility of tumor contouring: What was the Kappa for the two radiologists in defining solid and necrotic/cystic ROIs? An alternate to manual contouring would be to objectively define tumor sub-volumes using the histogram-based approach proposed by Chenevert, T. L. et al. Comparison of voxel-wise and histogram analyses of glioma ADC maps for prediction of early therapeutic change. Tomography 5, 7-14, 2019. https://doi.org/10.18383/j.tom.2018.00049

Thank you very much for this question. Each radiologist contoured either the T2w and DCE MRI sequence or the DWI and R2* map. As they were acquired in different orientations, the segmentations are not directly comparable. However, we have implemented an objective, histogram-based as described in the publication by Chenevert et al and included the results of the following analyses in the supporting materials (S1 for methods and S2 for results). We found that a threshold of 1.25x10-3mm2/s, as suggested by Chenevert et al., resulted in the identical interpretation of our results. While numerically different, the median D0 of the manually sub-segmented viable tumour and the automatically thresholded tissue were highly correlated (r = 0.75, p < 0.001, Figure R1). Similarly, responding tumours continued to show a significantly greater reduction in D0 following treatment and longer progression-free survival was associated with a greater reduction in D0.

a

b

Figure R1: a) D0 is strongly correlated between the manual sub-segmentation of viable tumour tissue and the automatic selection of viable voxels based on a published threshold. b) D0 remains a significant predictor of progression-free survival after the automated selection of viable tumour.

When the same thresholded masks were transferred to the perfusion fraction maps, a strong correlation with the manual sub-segmentation was observed (r = 0.90, p < 0.001). However, the perfusion fraction remained non-discriminatory between responders and non-responders and was not associated with PFS.

Figure R2: The perfusion fraction is strongly correlated between the manual sub-segmentation of viable tumour tissue and the automatic selection of viable voxels based on a published threshold for co-registered D0 maps.

No threshold for the differentiation of viable tumour and necrosis is available for Ktrans. We therefore chose to threshold on the goodness of fit of the extended Tofts model with a threshold of R2 = 0.5 to select only voxels which are perfused well enough to obtain a reasonable estimate of the kinetic parameters. We hypothesized that this would exclude necrotic and cystic tumour components. The Ktrans obtained through manual sub-segmentation of the tumour and the thresholding approach described above were strongly correlated (r = 0.69, p < 0.001). Furthermore, Ktrans remained non-discriminatory between responding and non-responding lesions and was not associated with survival.

Figure R3: Ktrans is strongly correlated between the manual sub-segmentation of viable tumour tissue and the automatic selection of viable voxels.

In the absence of a published threshold for the differentiation of viable tissue and necrosis based on R2*, we have employed the 90th percentile as an objective parameter in our analysis. Higher R2* values represent hypoxic tissue components. Similar to the median R2* in the viable tumour, the 90th percentile of the entire lesion increased significantly following treatment from 37 to 55 Hz (p < 0.001). However, neither the baseline value of R2* nor its change was associated with the RECIST response or survival. A comparison between the manual sub-segmentation and the thresholding at the 90th percentile showed a strong correlation between the two (r = 0.94, p < 0.001).

Figure R4: R2* is strongly correlated between the manual sub-segmentation of viable tumour tissue and the automatic selection of viable voxels.

2) Reproducibility of DCE-MRI: Reference 21 pertains to brain MRI, which may be much more reproducible than MRI of abdominal tumors. And reference 22 pertains to reproducibility of mouse DCE-MRI, which doesn’t seem too relevant to clinical DCE-MRI. I would suggest using the clinical DCE-MRI reproducibility figures published by Galbraith et al. (NMR Biomed 2002 Apr;15(2):132-42. doi: 10.1002/nbm.731.) and Klaassen et al. (Magnetic Resonance Imaging 50:1-9, 2018, https://doi.org/10.1016/j.mri.2018.02.005).

Thank you very much for these suggestions. We have replaced the references accordingly. As the study by Klaassen et al. was published after the NeoSun trial had started, we have adapted the wording to reflect that Klaassen et al. could not have informed the choice of the threshold in NeoSun. (Line 250)

3) Reproducibility of all quantitative MRI: In Figure 3, please also provide comparisons to corresponding changes in MRI parameters pre/post drug in tissues that are presumed to not be affected by drug. This is an option when repeat intra-session scanning was not done as part of the study. See, for example, Lorza, A.M.A., Ravi, H., Philip, R.C. et al. Dose–response assessment by quantitative MRI in a phase 1 clinical study of the anti-cancer vascular disrupting agent crolibulin. Sci Rep 10, 14449 (2020). https://doi.org/10.1038/s41598-020-71246-w.

Thank you very much for this suggestion. We have included normal renal tissue in our analysis, similar to the method described by Lorza et.al. in their publication. The diffusivity D0 and the perfusion fraction were unchanged in the normal kidney following 12 days of sunitinib (p = 0.52 and 0.63, respectively). R2* was borderline significantly increased at the follow-up (p = 0.05) but the change was significantly lower than the one observed in tumours (p < 0.001). Finally, KTrans was significantly increased in the normal kidney following treatment (p = 0.03) while the transfer constant was decreased in tumour tissue. In summary, the test -re-test variability in the normal kidney does not fundamentally alter the interpretation of changes observed in tumours following treatment or differences between responders and non-responders.

Revised Figure 3: Percentage change in the whole tumour volume (a), solid tumour volume (b), solid tumour Diffusivity D (c), solid tumour Ktrans (d), solid tumour R2* (e) and solid tumour perfusion fraction (f) between patients with stable/progressive disease (SD/PD) and partial/complete response (PR/CR). P-values for Mann-Whitney U test. Relative changes in normal renal tissue illustrate the test re-test variability in imaging measurements.

We have chosen normal renal tissue for the assessment of the repeatability of imaging measures as it was included in the field-of-view regardless of the orientation of the image acquisition. Furthermore, it had more similar contrast-enhancement characteristics to the tumours than skeletal muscle and, unlike the liver, had a single blood supply. Even though sunitinib is considered a targeted treatment for clear cell renal cell carcinoma, it binds multiple targets including vascular endothelial growth-factor receptors, platelet-derived growth-factor receptors, c-KIT, the RET proto-oncogene, the granulocyte colony-stimulating factor receptor and CD135. The range of on- and off-target adverse effects of sunitinib highlight the broad spectrum of tissue in which the drug exerts an effect. Therefore, it is challenging to establish whether any normal tissue is unaffected by the drug.

4) Univariable analysis: Median values of Ktrans, iAUC90, R2*, fp and D within the whole and solid tumor components were analyzed. Was T1 (unenhanced) also evaluated in the same manner?

Thank you very much for this question. T1 maps were acquired as described in the supporting materials (S1). However, their analysis was not pre-specified in the statistical analysis plan of the NeoSun trial. A post-hoc analysis showed a median tumour T1 of 1238 ms (interquartile range: 1184 -1341) at baseline which reduced to 1102 ms (994 – 1221 ms) after 12 days of sunitinib. The median reduction by 6% was not significant (p = 0.094). Neither the tumour T1 at baseline nor its change following treatment was associated with the RECIST response or survival. (Line 239)

5) It appears that there were secondary lesions that were large enough for RECIST assessment. What were the post-sunitinib vs. pre-sunitinib quantitative MRI results (analogous to figure 3) from those additional tumors?

Thank you for this question. The RECIST assessment was performed on contrast enhanced CT of the chest, abdomen and pelvis which covered an extended field of view compared to the multiparametric MRI. The most frequent sites of metastasis of renal cancer are the lungs and bone. In this study, 10/12 patients harboured lung metastasis, four patients presented with enlarged mediastinal lymph nodes, one patient with pelvic bone metastasis and one with enlarged inguinal lymph nodes which were not covered by the field of view. One patient presented with a contralateral adrenal metastasis, one with a retrocaval lymph node metastasis and one with a metastasis in the pancreatic head which were not covered by the field of view in the sagittal acquisitions. We have included this clarification in the methods section of the manuscript. (Line 167)

6) Discussion: Regarding the observed decrease in ADC of solid tumor following sunitinib treatment, is it known that the short-term response of tumor cells to sunitinib is cell swelling rather than apoptosis?

Thank you for this question. There is only little information on very early physiological response mechanisms of clear cell renal cell carcinoma to sunitinib. The only study investigating sequential changes in ADC in this setting found decreasing diffusion restriction after three days of treatment which increased again after 10 days 3. The authors have also hypothesized that changing contributions from the development of oedema, cell swelling and necrosis may contribute to the temporal changes in tumour ADC. After 18 weeks, the ADC is unchanged relative to the baseline imaging in another study 4. Furthermore, sunitinib also increases tumour immune cell infiltration which may also increase the diffusion restriction 5. We have expanded on this in the discussion. (Line 319)

7) Did the authors perform a multivariable analysis of whether baseline R2*, fp , D, T1 (unenhanced), Ktrans, and iAUC90, taken together (or a subset), are predictive of post- vs. pre-sunitinib change in Ktrans or iAUC90? A change in Tofts model parameters with treatment would be more mechanistically-related to tumor-level drug pharmacodynamics than RECIST, which is a downstream patient-level response metric. Use of delta-DCEMRI rather than RECIST as ground truth for outcome would also allow for the primary plus any metastatic tumors to be analyzed, which would increase sample size. The input data could be augmented to include not just the median values of R2*, fp , D, T1 (unenhanced), Ktrans, and iAUC90, but also additional percentile values (e.g., 10th and 90th) to both improve the predictive performance of the model and to try and do away with the need to manually define “solid” and “necrotic/cystic” tumor regions. Even of the outcome metric remained RECIST, a multivariable analysis would still be useful.

Thank you very much for this question.

We agree with the reviewer that the inclusion of objective parameters in addition to the manual sub-segmentation specified in the trial protocol would be useful. Therefore, we have included the threshold approach described by Klaassen et al. for the D0 and perfusion fraction maps. We have furthermore derived objective Ktrans measures through thresholding of the goodness of fit of the Toft’s model at R2 > 0.50, this only included voxels in the final analysis where kinetic parameters could be estimated with reasonable certainty and excluded poorly enhancing cystic and necrotic areas. The 90th percentile of the R2* distribution was selected as no threshold to distinguish viable and necrotic tissue has been published. The 90th percentile represents the most hypoxic tissue components. Please refer to question 3 of reviewer 2 for the results of these analyses.

We have not assessed whether any of the parameters was predictive of the reduction in Ktrans or iAUC90 because this was not specified in the trials statistical analysis plan. A post-hoc analysis showed that only the baseline Ktrans and iAUC90 were predictive of the relative reduction (p = 0.02 and 0.008, respectively). A higher Ktrans and iAUC90 at baseline were predictive of a greater reduction in Ktrans (R2adj = 0.35 and 0.64, respectively). Response prediction aims to anticipate patient benefit. Prolonged progression-free survival and RECIST are widely used surrogate markers of treatment benefit. The relationship between the pharmacodynamically relevant reduction in Ktrans and patient outcome is less well established. We have therefore chosen to summarise these results in the supplementary materials.

We have not performed a multi-variate analysis of all or a combination of the parameters. The reason was that the study was not powered for the development of such a multi-factorial model. As described in response to comment 5, metastatic deposits were outside of the field of view of the MRI acquisition and could unfortunately not serve to increase the sample size. Furthermore, using multiple lesions from one patient would introduce dependencies among data points which may lead to a less generalizable model. Overall, we felt that a population of 12 patients was not sufficient to explain a multifactorial model.

Thank you for the opportunity to review your paper on MRI results from this nice study.

Natarajan Raghunand, PhD

References

1. Wang HY, Su ZH, Xu X, et al. Dynamic Contrast-enhanced MR Imaging in Renal Cell Carcinoma: Reproducibility of Histogram Analysis on Pharmacokinetic Parameters. Sci Rep. 2016;6. doi:10.1038/srep29146

2. Wang H, Su Z, Ye H, et al. Reproducibility of Dynamic Contrast-Enhanced MRI in Renal Cell Carcinoma. Medicine (Baltimore). 2015;94(37):e1529. doi:10.1097/MD.0000000000001529

3. Desar IME, Voert EGW ter, Hambrock T, et al. Functional MRI techniques demonstrate early vascular changes in renal cell cancer patients treated with sunitinib: a pilot study. Cancer Imaging. 2011;11(1):259. doi:10.1102/1470-7330.2011.0032

4. Bharwani N, Miquel ME, Powles T, et al. Diffusion-weighted and multiphase contrast-enhanced MRI as surrogate markers of response to neoadjuvant sunitinib in metastatic renal cell carcinoma. Br J Cancer. 2014;110(3):616-624. doi:10.1038/bjc.2013.790

5. Haywood S, Chen R, Pavicic P, et al. Sunitinib’s effect on tumor infiltration of CD8 T cells in renal cell carcinoma (RCC) and modulation of their function by altering VEGF-induced upregulation of PD1 expression. https://doi.org/101200/jco2016342_suppl591. 2016;34(2_suppl):591-591. doi:10.1200/JCO.2016.34.2_SUPPL.591

Submitted filename: Reviewer Comments_V2.docx

Click here for additional data file.

11 Oct 2021

Multiparametric MRI for assessment of early response to neoadjuvant sunitinib in renal cell carcinoma

PONE-D-21-15997R1

Dear Dr. Barrett,

We’re pleased to inform you that your manuscript has been judged scientifically suitable for publication and will be formally accepted for publication once it meets all outstanding technical requirements.

Within one week, you’ll receive an e-mail detailing the required amendments. When these have been addressed, you’ll receive a formal acceptance letter and your manuscript will be scheduled for publication.

An invoice for payment will follow shortly after the formal acceptance. To ensure an efficient process, please log into Editorial Manager at http://www.editorialmanager.com/pone/, click the 'Update My Information' link at the top of the page, and double check that your user information is up-to-date. If you have any billing related questions, please contact our Author Billing department directly at authorbilling@plos.org.

If your institution or institutions have a press office, please notify them about your upcoming paper to help maximize its impact. If they’ll be preparing press materials, please inform our press team as soon as possible -- no later than 48 hours after receiving the formal acceptance. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information, please contact onepress@plos.org.

Kind regards,

Marco Giannelli

Academic Editor

PLOS ONE

Additional Editor Comments (optional):

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. If the authors have adequately addressed your comments raised in a previous round of review and you feel that this manuscript is now acceptable for publication, you may indicate that here to bypass the “Comments to the Author” section, enter your conflict of interest statement in the “Confidential to Editor” section, and submit your "Accept" recommendation.

Reviewer #1: All comments have been addressed

Reviewer #2: All comments have been addressed

**********

2. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Yes

Reviewer #2: Yes

**********

3. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: Yes

Reviewer #2: Yes

**********

4. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: Yes

Reviewer #2: Yes

**********

5. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

Reviewer #2: Yes

**********

6. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: The authors have addressed all the issues underlined by the reviewers. In my opinion the manuscript in this reviewed version it is worth for publication

Reviewer #2: (No Response)

**********

7. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

Reviewer #2: No

15 Oct 2021

PONE-D-21-15997R1

Multiparametric MRI for assessment of early response to neoadjuvant sunitinib in renal cell carcinoma

Dear Dr. Barrett:

I'm pleased to inform you that your manuscript has been deemed suitable for publication in PLOS ONE. Congratulations! Your manuscript is now with our production department.

If your institution or institutions have a press office, please let them know about your upcoming paper now to help maximize its impact. If they'll be preparing press materials, please inform our press team within the next 48 hours. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information please contact onepress@plos.org.

If we can help with anything else, please email us at plosone@plos.org.

Thank you for submitting your work to PLOS ONE and supporting open access.

Kind regards,

PLOS ONE Editorial Office Staff

on behalf of

Dr. Marco Giannelli

Academic Editor

PLOS ONE