Abstract
Background
Tumor perfusion may inform therapeutic response and resistance in metastatic renal cell carcinoma (RCC) treated with antiangiogenic therapy.
Purpose
To determine if arterial spin labeled (ASL) MRI perfusion changes are associated with tumor response and disease progression in metastatic RCC treated with vascular endothelial growth factor receptor (VEGFR) tyrosine kinase inhibitors (TKIs).
Materials and Methods
In this prospective study (ClinicalTrials.gov identifier: NCT00749320), metastatic RCC perfusion was measured with ASL MRI before and during sunitinib or pazopanib therapy between October 2008 and March 2014. Objective response rate (ORR) and progression-free survival (PFS) were calculated. Perfusion was compared between responders and nonresponders at baseline, at week 2, after cycle 2 (12 weeks), after cycle 4 (24 weeks), and at disease progression and compared with the ORR by using the Wilcoxon rank sum test and with PFS by using the log-rank test.
Results
Seventeen participants received sunitinib (mean age, 59 years ± 7.0 [standard deviation]; 11 men); 11 participants received pazopanib (mean age, 63 years ± 6.6; eight men). Responders had higher baseline tumor perfusion than nonresponders (mean, 404 mL/100 g/min ± 213 vs 199 mL/100 g/min ± 136; P = .02). Perfusion decreased from baseline to week 2 (−53 mL/100 g/min ± 31; P < .001), after cycle 2 (−65 mL/100 g/min ± 25; P < .001), and after cycle 4 (−79 mL/100 g/min ± 15; P = .008). Interval reduction in perfusion at those three time points was not associated with ORR (P = .63, .29, and .27, respectively) or PFS (P = .28, .27, and .32). Perfusion increased from cycle 4 to disease progression (51% ± 11; P < .001).
Conclusion
Arterial spin labeled perfusion MRI may assist in identifying responders to vascular endothelial growth factor receptor tyrosine kinase inhibitors and may help detect early evidence of disease progression in patients with metastatic renal cell carcinoma.
© RSNA, 2020
Online supplemental material is available for this article.
See also the editorial by Goh and De Vita in this issue.
Summary
Perfusion of renal cell carcinoma metastases, as quantified with arterial spin labeled MRI, may be useful in selecting candidates for antiangiogenic therapy and in helping detect the onset of treatment resistance.
Key Results
■ Responders to sunitinib or pazopanib had higher baseline perfusion (mean, 404 mL/100 g/min ± 213 [standard deviation]) than nonresponders (mean, 199 mL/100 g/min ± 136) (P = .02).
■ Mean perfusion increased at therapeutic resistance compared with mean perfusion after cycle 2 (36% ± 15; P = .03) and cycle 4 (53% ± 17; P = .006) (6 weeks per cycle).
■ Reduced mean perfusion from baseline to week 2 (−53 mL/100 g/min ± 31), cycle 2 (−65 mL/100 g/min ± 25), and cycle 4 (−79 mL/100 g/min ± 15) was not associated with response (P = .63, .29, and .27, respectively) or progression-free survival (P = .28, .27, and .32).
Introduction
Antiangiogenic therapies have revolutionized the management of metastatic renal cell carcinoma (RCC) in the past 15 years. Although front-line therapy has evolved with the introduction of immunotherapy, vascular endothelial growth factor receptor (VEGFR) tyrosine kinase inhibitors (TKIs) continue to play a major role in metastatic RCC management, either as a first-line therapy in combination with immunotherapy, as a second-line standalone therapy, or in patients with favorable-risk disease (1,2). These drugs achieve antitumor effect primarily through inhibition of vascular endothelial growth factor–driven angiogenesis (3–6). Initial response is common (7,8) but seldom durable or complete (4), and tumors develop resistance (9).
Effective VEGFR-TKI treatment may result in tumor stabilization without shrinkage (10–12). Thus, traditional therapy tracking using size-based measurements, such as Response Evaluation Criteria in Solid Tumors (RECIST) 1.1, may not be useful (13,14). Instead, imaging of tumor vascularity may serve as a marker of both treatment efficacy and failure (15–18).
Arterial spin labeled (ASL) MRI is a noninvasive technique that quantifies tissue perfusion. The spins of the endogenous water in inflowing arterial blood are magnetically labeled, resulting in signal intensity changes that are directly proportional to blood flow. Thus, ASL MRI is potentially more reliable than contrast material–enhanced techniques, which require assumptions about tissue permeability and contrast material input (19). Furthermore, ASL MRI does not require exogenous contrast material, which benefits patients with renal impairment or allergies (15). ASL MRI has been validated in both animals and humans (20–22).
In a previous study, ASL MRI evaluation of sorafenib-treated human renal cell carcinoma xenografts confirmed a correlation between ASL MRI signal intensity and tumor angiogenesis, with decreased perfusion observed after initiation of effective therapy and increased perfusion at treatment resistance. Low baseline perfusion helped predict poor sorafenib response, suggesting that ASL MRI–measured perfusion may serve as a marker for target engagement in vivo (21). In another study, ASL MRI showed that a decrease in tumor perfusion 1 month after initiation of vatalanib in patients with metastatic RCC correlated with improved prognosis at 4 months (22). However, the association between baseline perfusion and treatment response or between perfusion and onset of therapeutic resistance, which we hypothesize here, was not studied.
In this study, we used ASL MRI to assess tumor vascularity of renal cell carcinoma metastases before and during antiangiogenic therapy. The purpose was as follows: (a) to determine if baseline tumor perfusion and early changes in tumor perfusion after initiation of therapy predict response or progression-free survival (PFS) and (b) to associate changes in tumor perfusion throughout therapy with response and failure.
Materials and Methods
Participant Selection
This prospective study (ClinicalTrials.gov identifier: NCT00749320) was approved by institutional review boards at Beth Israel Deaconess Medical Center and the Dana Farber Cancer Institute and was compliant with the Health Insurance Portability and Accountability Act. The study was funded by Pfizer (New York, NY), GlaxoSmithKline (Brentford, United Kingdom), and Novartis (Basel, Switzerland). The authors had control of all data. Data generated or analyzed during the study are available from the corresponding author by request.
Written informed consent was obtained at both sites from consecutive, eligible agreeing participants from October 2008 to March 2014. Inclusion criteria were participants with metastatic RCC scheduled to receive either sunitinib or pazopanib and participants with at least one torso metastasis 2.5 cm or larger in diameter. Exclusion criteria were as follows: (a) participants received antiangiogenic therapy within 1 month of sunitinib and/or pazopanib initiation, (b) participants had MRI safety contraindications, (c) participants had a history of claustrophobia, and (d) there were no response data because of death or noncompliance.
VEGFR-TKI Therapy and End Points
Participants receiving sunitinib (Pfizer) and pazopanib (Novartis) were treated as first-line therapy with approved targeted doses of 50 mg daily for the first 4 of 6 weeks, or 800 mg daily continuously, respectively, in 6-week cycles. Response per patient was assessed by the treating oncologists for the first four cycles, then every 12 weeks, using CT results and RECIST 1.1. Further details are in Appendix E1 (online).
The end points were objective response rate (ORR) and PFS. ORR was the portion of participants achieving complete or partial response. PFS was the time from therapy initiation to the date of the CT examination documenting disease progression. Participants alive and without progression were censored at the last assessment.
MRI Protocol and Image Analysis
Participants underwent imaging at up to five time points: baseline, 2 weeks following therapy initiation, after cycles 2 and 4 of therapy, and at disease progression. MRI was performed at 1.5 T (TwinSpeed; GE Healthcare, Waukesha, Wis). Axial and coronal T2-weighted reference images were obtained, followed by ASL MRI of the target lesion(s) in one to three planes, depending on the visibility and anatomic location of the lesion. Acquisition parameters are provided in Table E1 (online). ASL MRI was performed with 1.5-second pseudocontinuous labeling and 1.5-second postlabeling delay. A proton density image, two inversion-recovery images for T1 quantification, and 16 pairs of label and control acquisitions were acquired. Perfusion maps were generated as previously described (23,24).
ASL MRI target selection and measurements were performed by one of two MRI fellowship–trained radiologists (I.P. and M.R.M.S, with 7 and 5 years of experience, respectively) who were blinded to all clinical data. Previous CT examinations were screened to identify optimal baseline ASL MRI target lesions in the torso. ASL MRI acquisition was then attempted at all subsequent time points. If there was more than one target lesion initially, only the lesion successfully imaged on all MRI scans was included for analysis (ie, one target lesion per participant). Because radiology reviewers were blinded to response assessment results, lesions chosen for ASL MRI were not necessarily RECIST target lesions.
ASL MRI perfusion measurements were obtained by drawing a freeform region of interest around the margin of the tumor on the proton density images and propagating this region of interest to the perfusion maps (Fig E1 [online]). The mean perfusion value was recorded. The imaging plane in which the lesion was best visualized at baseline was selected for analysis for all subsequent measurements.
Statistical Analysis
R software (version 3.6.2; R Foundation for Statistical Computing, Vienna, Austria) was used for statistical analysis. Changes in ASL MRI–measured perfusion relative to baseline during each time point were calculated as follows: %change in ASL(t) = 100 × (ASL(t) – baseline ASL)/baseline ASL, where ASL(t) is the perfusion at each treatment time point. Time points labeled “early” were those at 2 weeks and cycle 2. Change in perfusion was compared between responders and nonresponders and compared with the ORR by using the Wilcoxon rank sum test and with PFS (splitting percentage change at the median) by using the log-rank test. An optimal baseline perfusion cutoff value for responders was calculated by using the Youden index according to the Fisher exact test.
The percentage of change in ASL MRI relative to baseline at 2 weeks, after cycle 2, after cycle 4, and at disease progression was compared as paired sets by using the sign-rank test, and P values were adjusted for multiple comparisons by using the Holm-Bonferroni method. Mean difference in perfusion according to time point was assessed by using a random intercept linear mixed model, with time points used as a fixed effect and patient identifier as the random effect.
The incremental change in perfusion between observed and preceding time points was calculated as follows: %incremental change in ASL(t) = 100 × (ASL(t) – ASL(t–1))/ASL(t–1), where ASL(t–1) is the perfusion of the tumor at the previous time point. These were calculated for week 2, cycles 2 and 4, and at disease progression.
Two-sided P values are reported, and P ≤ .05 was considered to indicate a statistically significant difference. A power calculation is included in Appendix E1 (online).
Data were analyzed by using sunitinib and pazopanib cohorts separately and combined. A clear cell renal cell carcinoma (ccRCC) cohort was also studied because this is the most common subtype and the one with which efficacy of VEGFR-TKIs was established.
Results
Participant Characteristics
Thirty-two participants were screened (20 in the sunitinib cohort and 12 in the pazopanib cohort). The pazopanib cohort study was closed early because of slow accrual. Four participants were excluded, and 17 in the sunitinib cohort (mean age, 59 years ± 7.0; 11 men) and 11 in the pazopanib cohort (mean age, 63 years ± 6.6; eight men) were included in the final analysis (Fig 1). Table 1 shows participant demographic information, distribution, and histopathologic findings of lesions for both cohorts, and Table E2 (online) shows the number of ASL MRI studies completed at each time point.
Figure 1:
Patient inclusion flowchart for sunitinib and pazopanib cohorts. ASL = arterial spin labeled.
Table 1:
Participant Demographic Information, Lesion Locations, and Histopathologic Findings
Within the sunitinib group, nine participants underwent ASL MRI measurements during cycle 2; seven of these nine participants underwent ASL MRI measurements the day after the end of the treatment phase. One participant delayed both the MRI examination and start of cycle 3 for 53 days because of scheduling issues and sunitinib intolerance, and the second participant delayed the MRI examination for 16 days because of travel. A third participant had one ASL MRI measurement performed on time, but the participant had stopped taking sunitinib 14 days before because of intolerance. The median time between the end of cycle 2 treatment and ASL MRI measurements across the nine participants was 1 day (mean, 9.9 days ± 17.3 [standard deviation]). For the six participants who underwent cycle 4 ASL MRI measurements, one participant underwent ASL MRI 8 days after the last dose because of stoppage from intolerance, and one delayed the MRI examination for 5 days because of scheduling issues; the median time between the end of treatment and ASL MRI in this group was 1 day (mean, 2.8 days ± 3.0). Two participants in the pazopanib cohort left the study because of adverse effects before any response assessment could be made.
Eight participants in the sunitinib cohort and four participants in the pazopanib cohort had two ASL MRI candidate lesions at baseline and none had three. Each participant had only one lesion with completed ASL MRI measurements at all follow-up MRI examinations. An example of tumor size and ASL MRI measurement change with therapy is shown in Figure 2.
Figure 2:
Images in 65-year-old man with metastatic renal cell carcinoma to the lung. T2-weighted axial single-shot fast spin-echo MRI scans (top row) and arterial spin labeled MRI perfusion maps (bottom row) of pulmonary metastasis show decreased tumor size (green caliper lines in top row correspond to long- and short-axis measurements) and perfusion (green regions of interest in bottom row; numbers on images are calculated flow), compared with baseline during treatment at week 2 and cycle 2, and enlargement and increased perfusion at disease progression, which occurred at cycle 4.
Treatment Response
Table 2 lists the distribution of best treatment responses and ORR for each cohort. No participants achieved complete response. Four participants in the sunitinib cohort and two participants in the pazopanib cohort achieved partial response as their best outcome. The median PFS was 7.8 months (95% CI: 2.3, 14.6) for the sunitinib cohort and 6.3 months (95% CI: 2.0, 8.8) for the pazopanib cohort.
Table 2:
Best Treatment Response and Objective Response Rate for Sunitinib and Pazopanib Cohorts
Treatment Response and Baseline Perfusion Levels
In the sunitinib cohort, participants who attained response had statistically higher basline ASL MRI perfusion levels (median, 403 mL/100 g/min; mean, 459 mL/100 g/min ± 210) than nonresponders (median, 240 mL/100 g/min; mean, 207 mL/100 g/min ± 142) (P = .03) (Fig 3). This was not observed in the pazopanib cohort, where the median ASL MRI perfusion level was 294 mL/100 g/min (mean, 294 mL/100 g/min ± 241) for responders and 152 mL/100 g/min (mean, 187 mL/100 g/min ± 133) for nonresponders (P = .56). Baseline perfusion was not different between responders in the sunitinib and pazopanib cohorts (P = .22). When both groups were combined, responders demonstrated a higher baseline perfusion than nonresponders (median, 387 mL/100 g/min [mean, 404 mL/100 g/min ± 213] vs 156 mL/100 g/min [mean,199 mL/100 g/min ± 136]; P = .02). This was also true when only ccRCCs were examined (median, 387 mL/100 g/min [mean, 390 mL/100 g/min ± 105] for responders vs 200 mL/100 g/min [mean, 206 mL/100 g/min ± 122] for nonresponders; P = .02).
Figure 3:
Box-and-whisker plots of baseline arterial spin labeled (ASL) MRI–measured tumor perfusion for nonresponders versus responders in different cohorts. Shaded box represents interquartile range (25%–75%), bold line represents median, and whiskers represent overall range. Dots represent individual data points. ASL MRI perfusion is reported in milliliters per 100 grams per minute. ccRCC = clear cell renal cell carcinoma, ORR = objective response rate.
Association between Treatment Response and Cutoff Value
An optimal threshold of 290 mL/100 g/min to segregate responders from nonresponders was calculated with the Youden index. An association existed between treatment response and cutoff value for the combined (P = .02), sunitinib (P = .03), and ccRCC (P = .01) cohorts but not for the pazopanib cohort (P = .49) (Table E3 [online]). A total of 11 participants (eight treated with sunitinib and three treated with pazopanib) met this threshold, with a sensitivity of 83% (five of six participants; 95% CI: 36, 99.6), a specificity of 73% (16 of 22 participants; 95% CI: 50, 89), a positive predictive value of 45% (five of 11 participants; 95% CI: 28, 64), and a negative predictive value of 94% (16 of 17 participants; 95% CI: 72, 99) for the combined cohort.
Percentage Change in Perfusion Compared with Baseline
The percentage of changes in perfusion compared with baseline at 2 weeks, cycles 2 and 4, and at disease progression are shown in Figure 4. There was a decrease in tumor perfusion in the early treatment phase in the sunitinib cohort and for the combined cohort (baseline to week 2 and cycle 2) and for the pazopanib cohort (baseline to week 2) (Table 3). However, no differences in early perfusion change were seen at week 2 or cycle 2 between responders and nonresponders for sunitinib (P > .99 and = .52) or pazopanib (P = undefined with only one responder at week 2, .77) (Fig E2 [online]), in the combined cohort (P = .60 and .29), or in ccRCC (P = .20 and .08). Early perfusion changes, when split at the median value into “high” and “low” groups, also did not show differences in PFS for individual (sunitinib, P = .33; pazopanib, P = .47) or combined (P = .28) cohorts or for ccRCCs (P = .16) (Fig E3 [online]). No difference existed in PFS when using the optimized perfusion cutoff of 290 mL/100 g/min for the sunitinib (P = .99), pazopanib (P = .62), combined (P = .61), or ccRCC (P = .49) cohorts. Among the nine total participants who underwent ASL measurements at cycle 4 (three treated with pazopanib and six treated with sunitinib), no association existed between ASL-measured tumor perfusion and treatment response (P = .27) or PFS (P = .32).
Figure 4:
Box-and-whisker plots of percentage change in arterial spin labeled (ASL) MRI–measured tumor perfusion in comparison to baseline at each time point in different cohorts. Shaded box represents interquartile range (25%–75%), bold line represents median, and whiskers represent range as defined by 1.5 × interquartile range or maximum/minimum closer to median, whichever is smaller. Dots represent individual data points. ASL MRI perfusion reported in milliliters per 100 grams per minute. ccRCC = clear cell renal cell carcinoma, Dx Prog. = diagnosis of progression.
Table 3:
Percentage of Changes in ASL MRI–measured Perfusion at Each Time Point Compared with Baseline
Estimated Differences in Percentage Perfusion Change from Baseline at Each Time Point
Table 4 shows the linear mixed model used to provide an estimated difference in percentage of perfusion change from baseline at each time point in the sunitinib cohort, the combined cohorts, and for ccRCC, using the disease progression time point as a reference. In the sunitinib cohort, differences existed between the percentage of perfusion change from cycle 2 to disease progression (mean ± standard error, 35.6% ± 15.08; P = .03) and from cycle 4 to disease progression (mean, 52.5% ± 16.85; P = .006). When the cohorts were combined or when only ccRCCs were analyzed, there were differences between the percentage of perfusion change at all time points when compared with disease progression (Table 4). A similar calculation could not be performed for the pazopanib cohort because of limited sample size. When comparing incremental changes in perfusion, a similar pattern of relatively increased perfusion was seen between cycle 4 and disease progression, but there was insufficient sample size to assess for statistical significance (Fig E4 [online]).
Table 4:
Estimated Difference in Average Percentage of Change in ASL MRI Perfusion from Baseline for Each Time Point within the Sunitinib, Combined, and ccRCC Cohorts
Discussion
In this prospective clinical trial, we used arterial spin labeled (ASL) MRI to measure perfusion of renal cell carcinoma (RCC) metastases treated with vascular endothelial growth factor receptor (VEGFR) tyrosine kinase inhibitors (TKIs) to determine if baseline tumor perfusion or changes in perfusion during treatment were associated with response or progression-free survival (PFS). Higher baseline perfusion was associated with objective response rate in the sunitinib, combined, and clear cell RCC (ccRCC) cohorts (P = .03, .02, and .02, respectively), and a threshold of 290 mL/100 g/min could help differentiate responders from nonresponders (P = .03, .02, and .01). These cohorts also demonstrated higher perfusion at disease progression when compared with perfusion at 12 weeks and 24 weeks of treatment (P < .001 for all). Baseline and interval changes in perfusion were not associated with PFS (P = .16–.99).
ASL MRI may offer an alternative to RECIST-based measures of metastatic RCC response by providing an objective, quantitative marker of likelihood of response to VEGFR-TKIs. Our results are concordant with a study using human renal cell carcinoma xenografts (Caki-1, 786-0, and A498) treated with sorafenib, where baseline perfusion values correlated with therapeutic response (21). Our results suggest that baseline tumor perfusion could play a role in metastatic RCC management when VEGFR-TKIs are considered.
We observed a decrease in tumor perfusion within the early treatment period in both cohorts, reflective of the direct antiangiogenic action of VEGFR-TKIs as confirmed in xenograft-based studies (21,25). This is consistent with a recent study in which a 29% decrease in ASL MRI perfusion was detected in patients with metastatic RCC receiving anti-hypoxia inducible factor 2 (HIF-2) therapy (26). However, therapeutic changes in tumor perfusion were not predictive of response in either study.
In our study, the degree of change in ASL MRI–measured perfusion was not associated with PFS, in contrast to a previous study with vatalanib, which showed an association (22). However, four of 10 patients in the vatalanib study showed a net increase in tumor perfusion within 4 weeks of treatment, whereas only one of 26 participants who had undergone imaging at week 2 and cycle 2 did so in our study. This high rate of early response is typical for sunitinib and pazopanib but not for vatalanib. Also, the end point for the vatalanib study was defined at 4 months, so patients who showed clinical progression afterward would have been considered responders; indeed, our median PFS in both cohorts was longer (7.9 months and 6.3 months, respectively). Instead, our end point was defined as the time of disease progression, with follow-up occurring up to 41 months.
An increase in ASL MRI–measured perfusion was seen at disease progression when compared with measurements during treatment in the sunitinib, combined, and ccRCC cohorts. This is concordant with xenograft renal cell carcinoma models (25,27), suggesting that perfusion measurements may be useful in determining treatment resistance by directly detecting angiogenic escape. However, the ability to use perfusion imaging to help detect resistance before RECIST-based determination was not realized in our study, as the detected perfusion increase occurred only at the time of progression as determined with RECIST criteria and not before.
Our study had several limitations, the main one being the early study termination, which resulted in less-than-intended enrollment, possibly preventing the detection of an association between perfusion and PFS. The pazopanib cohort was the smallest and underpowered, although the results paralleled findings from the sunitinib and combined cohorts. Also, eight of 28 participants did not have ccRCC, a higher percentage than the typical metastatic RCC demographic group, reflecting selection bias. Because non-ccRCCs are known to have a relatively lower response rate to sunitinib or pazopanib, some analyses may have been skewed. We believe this did not affect our findings because our analysis using only patients with ccRCC had the same results as the combined, sunitinib, and, when sufficiently powered, pazopanib cohorts. The lesions selected for ASL MRI were also not necessarily matched to those used in RECIST, preventing a direct comparison. Also, we did not include tests of ASL MRI reproducibility; however, a previous study measuring normal human renal perfusion was performed with the same MRI scanner with an identical ASL MRI acquisition protocol, yielding a test-retest reproducibility of 7% (23). Finally, ASL MRI was not always performed immediately after sunitinib treatment in cycles 2 and 4 as planned, and there may have been recovery of tumor perfusion during this gap, which may have obscured perfusion differences.
In conclusion, we showed pretreatment arterial spin labeled (ASL) MRI–measured tumor perfusion in participants with metastatic renal cell carcinoma may inform about the likelihood of antiangiogenic therapy response, and ASL MRI may be used to track treatment response and failure. This may be used to inform management for metastatic renal cell carcinoma, where the two current first-line therapies in the United States are a combination of two immunotherapies or a combination of vascular endothelial growth factor receptor tyrosine kinase inhibitors plus immunotherapy. On the basis of our results, patients with low tumor perfusion at baseline may benefit more by initiating the former. Future ASL MRI–based studies, both with a larger cohort size and in combination with immunotherapy, and incorporating reproducibility measures, would be of interest because angiogenic escape mechanisms may differ and occur at different time points relative to Response Evaluation Criteria in Solid Tumors–based detection. The results of these studies may help improve the performance of ASL MRI in helping predict therapeutic outcomes.
APPENDIX
SUPPLEMENTAL FIGURES
Disclosures of Conflicts of Interest: L.L.T. Activities related to the present article: institution received grants from Pfizer, GlaxoSmithKline, and Novartis. Activities not related to the present article: is a consultant for Agile Devices; receives royalties from UpToDate; holds stock/stock options in many entities. Other relationships: disclosed no relevant relationships. R.S.B. Activities related to the present article: institution received grant from National Institutes of Health (NIH). Activities not related to the present article: disclosed no relevant relationships. Other relationships: disclosed no relevant relationships. M.F.S. Activities related to the present article: institution received grants from NIH, Pfizer, and GlaxoSmithKline; institution received payment from Pfizer and Novartis for writing or reviewing the manuscript. Activities not related to the present article: disclosed no relevant relationships. Other relationships: disclosed no relevant relationships. O.A.J. disclosed no relevant relationships. M.R.M.S. Activities related to the present article: institution received grants from Pfizer, GlaxoSmithKline, and Novartis. Activities not related to the present article: disclosed no relevant relationships. Other relationships: disclosed no relevant relationships. D.C.A. Activities related to the present article: institution received grants from Pfizer, Novartis, and GlaxoSmithKline. Also receives royalties for arterial spin labeled–related patents from GE, Siemens, Philips, Hitachi, Animage Technologies, and UIH America. Activities not related to the present article: institution has grants/grants pending with GE Healthcare. Other relationships: disclosed no relevant relationships. P.C. Activities related to the present article: institution received grant from the National Cancer Institute. Activities not related to the present article: disclosed no relevant relationships. Other relationships: disclosed no relevant relationships. D.M. disclosed no relevant relationships. P.M.R. disclosed no relevant relationships. M.B.A. Activities related to the present article: institution received grants from Pfizer, Novartis, and GlaxoSmithKline. Activities not related to the present article: receives payment for board membership from Pyxis Oncology, Werewolf, and Leads Oncology; is a consultant for Pfizer, Novartis, Exelixis, BMS, Merck, Aveo, Eisa, Genentech, Agenus, Apexigen, and Arrowhead; receives payment from Werewolf and Pyxis Oncology for development of educational presentations; receives payment from UpTodate, Cowen, Leerink, and Cota. Other relationships: disclosed no relevant relationships. I.P. Activities related to the present article: disclosed no relevant relationships. Activities not related to the present article: is a consultant for Bayer Healthcare; has patents planned, pending, or issued with Philips Healthcare; holds stock/stock options in Health Tech International. Other relationships: has patent pending with University of Texas Southwestern Medical Center.
This study was supported by National Institutes of Health (NIH) Kidney Cancer Specialized Program of Research Excellence grants (P50CA101942, P50CA196516). The sunitinib cohort of the study was funded by a grant from Pfizer (WS277615). The pazopanib cohort was funded by GlaxoSmithKline and Novartis (CPZP034AUS53T). I.P. was supported by grants from NIH (2R01CA154475, U01CA207091).
Abbreviations:
- ASL
- arterial spin labeled
- ccRCC
- clear cell RCC
- ORR
- objective response rate
- PFS
- progression-free survival
- RCC
- renal cell carcinoma
- RECIST
- Response Evaluation Criteria in Solid Tumors
- TKI
- tyrosine kinase inhibitor
- VEGFR
- vascular endothelial growth factor receptor
References
- 1.Motzer RJ, Tannir NM, McDermott DF, et al. Nivolumab plus Ipilimumab versus Sunitinib in Advanced Renal-Cell Carcinoma. N Engl J Med 2018;378(14):1277–1290. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Shah AY, Kotecha RR, Lemke EA, et al. Outcomes of patients with metastatic clear-cell renal cell carcinoma treated with second-line VEGFR-TKI after first-line immune checkpoint inhibitors. Eur J Cancer 2019;114:67–75. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Chouhan JD, Zamarripa DE, Lai PH, Oramasionwu CU, Grabinski JL. Sunitinib (Sutent): a novel agent for the treatment of metastatic renal cell carcinoma. J Oncol Pharm Pract 2007;13(1):5–15. [DOI] [PubMed] [Google Scholar]
- 4.Joosten SC, Hamming L, Soetekouw PM, et al. Resistance to sunitinib in renal cell carcinoma: From molecular mechanisms to predictive markers and future perspectives. Biochim Biophys Acta 2015;1855(1):1–16. [DOI] [PubMed] [Google Scholar]
- 5.Minardi D, Quaresima L, Santoni M, et al. Recent aspects of sunitinib therapy in patients with metastatic clear-cell renal cell carcinoma: a systematic review of the literature. Curr Urol Rep 2015;16(2):3. [DOI] [PubMed] [Google Scholar]
- 6.Motzer RJ, Hutson TE, Tomczak P, et al. Sunitinib versus interferon alfa in metastatic renal-cell carcinoma. N Engl J Med 2007;356(2):115–124. [DOI] [PubMed] [Google Scholar]
- 7.Motzer RJ, Hutson TE, Tomczak P, et al. Overall survival and updated results for sunitinib compared with interferon alfa in patients with metastatic renal cell carcinoma. J Clin Oncol 2009;27(22):3584–3590. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Motzer RJ, Rini BI, Bukowski RM, et al. Sunitinib in patients with metastatic renal cell carcinoma. JAMA 2006;295(21):2516–2524. [DOI] [PubMed] [Google Scholar]
- 9.Rini BI, Atkins MB. Resistance to targeted therapy in renal-cell carcinoma. Lancet Oncol 2009;10(10):992–1000. [DOI] [PubMed] [Google Scholar]
- 10.O’Reilly MS. The preclinical evaluation of angiogenesis inhibitors. Invest New Drugs 1997;15(1):5–13. [DOI] [PubMed] [Google Scholar]
- 11.Boehm T, Folkman J, Browder T, O’Reilly MS. Antiangiogenic therapy of experimental cancer does not induce acquired drug resistance. Nature 1997;390(6658):404–407. [DOI] [PubMed] [Google Scholar]
- 12.Harris AL. Antiangiogenesis for cancer therapy. Lancet 1997;349(Suppl 2):SII13–SII15. [DOI] [PubMed] [Google Scholar]
- 13.Eisenhauer EA, Therasse P, Bogaerts J, et al. New response evaluation criteria in solid tumours: revised RECIST guideline (version 1.1). Eur J Cancer 2009;45(2):228–247. [DOI] [PubMed] [Google Scholar]
- 14.Scott PA, Harris AL. Current approaches to targeting cancer using antiangiogenesis therapies. Cancer Treat Rev 1994;20(4):393–412. [DOI] [PubMed] [Google Scholar]
- 15.Pedrosa I, Alsop DC, Rofsky NM. Magnetic resonance imaging as a biomarker in renal cell carcinoma. Cancer 2009;115(10 Suppl):2334–2345. [DOI] [PubMed] [Google Scholar]
- 16.Kang HC, Tan KS, Keefe SM, et al. MRI assessment of early tumor response in metastatic renal cell carcinoma patients treated with sorafenib. AJR Am J Roentgenol 2013;200(1):120–126. [DOI] [PubMed] [Google Scholar]
- 17.Hahn OM, Yang C, Medved M, et al. Dynamic contrast-enhanced magnetic resonance imaging pharmacodynamic biomarker study of sorafenib in metastatic renal carcinoma. J Clin Oncol 2008;26(28):4572–4578. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Rosen MA, Schnall MD. Dynamic contrast-enhanced magnetic resonance imaging for assessing tumor vascularity and vascular effects of targeted therapies in renal cell carcinoma. Clin Cancer Res 2007;13(2 Pt 2):770s–776s. [DOI] [PubMed] [Google Scholar]
- 19.Williams DS, Detre JA, Leigh JS, Koretsky AP. Magnetic resonance imaging of perfusion using spin inversion of arterial water. Proc Natl Acad Sci U S A 1992;89(1):212–216. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Telischak NA, Detre JA, Zaharchuk G. Arterial spin labeling MRI: clinical applications in the brain. J Magn Reson Imaging 2015;41(5):1165–1180. [DOI] [PubMed] [Google Scholar]
- 21.Schor-Bardach R, Alsop DC, Pedrosa I, et al. Does arterial spin-labeling MR imaging-measured tumor perfusion correlate with renal cell cancer response to antiangiogenic therapy in a mouse model? Radiology 2009;251(3):731–742. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.de Bazelaire C, Alsop DC, George D, et al. Magnetic resonance imaging-measured blood flow change after antiangiogenic therapy with PTK787/ZK 222584 correlates with clinical outcome in metastatic renal cell carcinoma. Clin Cancer Res 2008;14(17):5548–5554. [DOI] [PubMed] [Google Scholar]
- 23.Robson PM, Madhuranthakam AJ, Dai W, Pedrosa I, Rofsky NM, Alsop DC. Strategies for reducing respiratory motion artifacts in renal perfusion imaging with arterial spin labeling. Magn Reson Med 2009;61(6):1374–1387. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Lanzman RS, Robson PM, Sun MR, et al. Arterial spin-labeling MR imaging of renal masses: correlation with histopathologic findings. Radiology 2012;265(3):799–808. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Bhatt RS, Wang X, Zhang L, et al. Renal cancer resistance to antiangiogenic therapy is delayed by restoration of angiostatic signaling. Mol Cancer Ther 2010;9(10):2793–2802. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Courtney KD, Ma Y, Diaz de Leon A, et al. HIF-2 Complex Dissociation, Target Inhibition, and Acquired Resistance with PT2385, a First-in-Class HIF-2 Inhibitor, in Patients with Clear Cell Renal Cell Carcinoma. Clin Cancer Res 2020;26(4):793–803. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Zhang L, Bhasin M, Schor-Bardach R, et al. Resistance of renal cell carcinoma to sorafenib is mediated by potentially reversible gene expression. PLoS One 2011;6(4):e19144. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.