Eur Radiol DOI 10.1007/s00330-015-3591-9
ONCOLOGY
Optimal follow-up intervals in active surveillance of renal masses in patients with von Hippel-Lindau disease Fabio Pomerri & Giuseppe Opocher & Chiara Dal Bosco & Pier Carlo Muzzio & Gisella Gennaro
Received: 12 May 2014 / Revised: 17 December 2014 / Accepted: 12 January 2015 # European Society of Radiology 2015
Abstract Objectives To estimate an optimal follow-up (FU) interval for von Hippel-Lindau (VHL) patients with renal masses (RMs) by determining tumour growth rates from growth curves. Methods Thirty lesions (47.6 %) were classified as solid tumours (STs) and 33 (52.4 %) as complex cysts (CCs). Variations in lesion volume over time were analyzed. For 53 lesions, we calculated the growth rate during the period when the volume of the lesion changed most rapidly, and called this the fast growth rate (FGR). Results The STs initially grew fast, followed by a period of slower growth. The CCs varied in volume over time, associated with variable amounts of their fluid component. The FGR correlated better with the latest volume for STs (r=0.905) than for CCs (r=0.780). An optimal FU interval between 3 and 12 months was derived by combining the FGR calculated from the curve with the latest volume measured. Conclusions Analyzing growth curves and related kinetic parameters for RMs in VHL patients could be useful with a view to optimizing the subsequent FU interval and improving the active surveillance program.
F. Pomerri : C. D. Bosco : P. C. Muzzio : G. Gennaro (*) Radiology Unit, Veneto Institute of Oncology IOV—IRCCS, via Gattamelata, 64, 35128 Padua, Italy e-mail: [email protected] G. Opocher Familial Cancer Clinics and Oncoendocrinology Unit, Veneto Institute of Oncology IOV—IRCCS, Padua, Italy F. Pomerri : G. Opocher Department of Medicine, University of Padua, Padua, Italy
Key Points • Measuring volume changes over time enables tumour growth curves to be charted. • Renal solid tumours increase in volume with a typical sigmoidal curve. • Complex cysts may increase and decrease in volume spontaneously over time. • The fast growth rate of solid tumours correlates with their latest volume. • The fast growth rate can orient the scheduling of subsequent follow-ups. Keywords von Hippel-Lindau disease . Growth curve . Renal mass . Active surveillance . Magnetic resonance imaging Abbreviations CCs Complex cysts DT Doubling time FGR Fast growth rate FU Follow-up GR Growth rate MRI Magnetic resonance imaging RCC Renal cell carcinoma RMs Renal masses STs Solid tumours VHL Von Hippel-Lindau
Introduction von Hippel-Lindau (VHL) disease is a rare, autosomal dominant, inherited neoplastic disorder characterized by the development of a variety of benign and malignant tumours, including retinal and central nervous system hemangioblastomas, renal cysts and tumours, and pancreatic cysts and tumours,
Eur Radiol
among others [1, 2]. The prevalence of VHL disease is estimated at one in 36,000 births, and its penetrance is over 90 % by the age of 65 [3]. It is caused by a mutation of the VHL gene located on chromosome 3p25 [4]. Renal cell carcinoma (RCC) is one of the most common causes of death in patients with VHL disease and is seen in 24–45 % of patients. When renal cysts are also considered, 60 % of VHL patients have renal lesions. RCC often remains asymptomatic for long periods of time, and serial imaging is needed for its early diagnosis [1, 5]. Complex cysts (CCs) are fluid cystic masses, often containing solid RCC components, and are usually treated as potentially malignant lesions. They need to be monitored, although transitions from CCs to solid lesions are quite rare [1, 6]. Renal masses (RMs) in VHL patients are often multiple and bilateral, and their growth rate varies considerably, complicating their management, but the general strategy is to delay surgical treatment for as long as possible while keeping the risk of metastases reasonably low. Active surveillance is needed to enable kidney function for as long as possible, to avoid dialysis, and limit the risk of invasive disease, with serial cross-sectional imaging to monitor a lesion’s growth until its size (maximum diameter) reaches a threshold conventionally set at 3 cm. This is because it has been demonstrated that RCCs less than 3 cm in diameter rarely metastasize [1, 7, 8]. When STs reach or exceed this 3-cm threshold, several surgical treatments are possible, including radical nephrectomy, nephron-sparing surgery, and the more recently introduced minimally-invasive techniques (radiofrequency ablation and cryoablation) [9–11]. There are no specific guidelines on the timing of the FU for VHL patients, but MRI imaging is usually scheduled once a year. Experimental cancer research on animals has demonstrated, however, that tumours grow exponentially only as long as they are small, while their growth rate slows down as they become larger (as in the Gompertzian or logistic models) [12–14]. In clinical studies, estimates of the rate of volumetric growth of untreated tumours are usually based on brief intervals, for which the exponential model can be applied to describe a tumour’s growth, and endpoints such as doubling time (DT) can be used to assess a tumour’s aggressiveness [15]. As mentioned earlier, tumours in VHL patients are usually left to grow untreated until they reach a given size threshold [16, 17], and a non-exponential model of growth - with its typical sigmoidal shape - is assumed to be applicable. Actual tumour growth can be checked in vivo by means of an active surveillance program for VHL patients, measuring the volume of their lesions by means of imaging at each FU. The purpose of the study was to estimate an optimal FU interval between consecutive MRI examinations in VHL patients with RMs, based on the assessment of growth curves and the determination of tumour growth rates.
Materials and methods Study population The study was approved by the Institutional Review Board; the need to obtain patients’ informed consent was waived, given the retrospective nature of the study. Forty-four consecutive patients with genotype-confirmed VHL disease were identified using the institutional electronic medical record system to obtain demographic information, relevant medical history, pertinent radiology, and pathology reports. Of these 44 VHL patients, 26 (59.1 %) had renal involvement and comprise the study population. These patients all had RMs, either solid tumours (STs), or CCs, or both, in one or both kidneys. Twelve of them had previously undergone surgery for RCC, involving either nephrectomy, or nephron-sparing surgery. Simple cysts [18] were not considered in this study. All patients had been under active surveillance with a periodical MRI for a total period ranging from 7 to 104 months (mean: 50.3 months) with variable FU intervals. In vivo RM growth curves were constructed and analyzed by determining the volume of 63 RMs in all MRIs obtained during the surveillance period. Imaging technique Active surveillance for VHL patients was based on MRI. All imaging performed at our institute was done with a 1.5 T unit (Magnetom Espree; Siemens Medical Systems, Erlangen, Germany) with phased-array coils (four-element anterior and eight-element posterior coil arrays) to increase the signal-to-noise ratio. Each series of images included the following: (1) axial and coronal breath-hold T2-weighted MRI with half-Fourier singleshot turbo spin-echo sequences; (2) axial in- and opposed-phase T1-weighted spoiled gradient-echo sequences; (3) axial three-dimensional fat-suppressed T1weighted interpolated spoiled gradient-echo sequences (volumetric interpolated breath-hold examination, VIBE) acquired before and during arterial, venous, and delayed phases, after intravenous administration of contrast medium (0.2 mL/kg up to a maximum of 16 mL; flow rate 2–2.5 mL/sec; Omniscan; GE Healthcare A.S., Oslo, Norway). Parameter settings for each MRI sequence are summarized in Table 1. Some MRI examinations performed at other hospitals (31 out of 162; 19.1 %) were included in the study because sequences had been obtained before and after injection of the contrast medium, and the images were of adequate diagnostic quality.
Eur Radiol Table 1 Parameter settings for each sequence used in MRI examinations for VHL patients included in the study Parameter
T2-weighted
T1-weighted
VIBE
Repetition time (msec) Echo time (msec)
1,500 93
3.8 1.4
Flip angle (°) Matrix (pixel) Section thickness (mm) Intersection gap (mm) Field of view (mm2)
170 180×320 6 1.2 400×300
129 4.9 in phase 2.4 out phase 70 162×256 6 1.2 380×308
10 150×320 2.3 0.5 420×301
Image analysis All MRI images were examined independently by two experienced radiologists (F.P., C.D.B.), who assessed the characteristics of all RMs, classifying them as STs or CCs, and measured three maximum diameters (two axial and one coronal) of each lesion to calculate their volume. If the lesion was classified in the same way by both radiologists and the differences between their measurements were small (1 cm3/month), suggesting that the FGR could be used to adjust the FU interval between two consecutive imaging procedures in VHL patients with a view to closely monitoring ST growth. In fact, the FGR provides a quantitative indication of a lesion’s growth and could help clinicians to schedule subsequent MRI examinations; an optimal FU interval should be established depending on the FGR, the latest-recorded volume of a given mass, and the overall stage of the growth curve. Table 3 suggests the FU interval associated with a given FGR and latest volume, assuming that the growth curve or a portion of it is known.
4
6
8
10
12
14
16
18
20
Vollatest (cm3)
The first condition to meet in order to use these criteria is to obtain enough information about the growth curve to be able to ascertain the FGR. If a new, small lesion is discovered (volume
ONCOLOGY
Optimal follow-up intervals in active surveillance of renal masses in patients with von Hippel-Lindau disease Fabio Pomerri & Giuseppe Opocher & Chiara Dal Bosco & Pier Carlo Muzzio & Gisella Gennaro
Received: 12 May 2014 / Revised: 17 December 2014 / Accepted: 12 January 2015 # European Society of Radiology 2015
Abstract Objectives To estimate an optimal follow-up (FU) interval for von Hippel-Lindau (VHL) patients with renal masses (RMs) by determining tumour growth rates from growth curves. Methods Thirty lesions (47.6 %) were classified as solid tumours (STs) and 33 (52.4 %) as complex cysts (CCs). Variations in lesion volume over time were analyzed. For 53 lesions, we calculated the growth rate during the period when the volume of the lesion changed most rapidly, and called this the fast growth rate (FGR). Results The STs initially grew fast, followed by a period of slower growth. The CCs varied in volume over time, associated with variable amounts of their fluid component. The FGR correlated better with the latest volume for STs (r=0.905) than for CCs (r=0.780). An optimal FU interval between 3 and 12 months was derived by combining the FGR calculated from the curve with the latest volume measured. Conclusions Analyzing growth curves and related kinetic parameters for RMs in VHL patients could be useful with a view to optimizing the subsequent FU interval and improving the active surveillance program.
F. Pomerri : C. D. Bosco : P. C. Muzzio : G. Gennaro (*) Radiology Unit, Veneto Institute of Oncology IOV—IRCCS, via Gattamelata, 64, 35128 Padua, Italy e-mail: [email protected] G. Opocher Familial Cancer Clinics and Oncoendocrinology Unit, Veneto Institute of Oncology IOV—IRCCS, Padua, Italy F. Pomerri : G. Opocher Department of Medicine, University of Padua, Padua, Italy
Key Points • Measuring volume changes over time enables tumour growth curves to be charted. • Renal solid tumours increase in volume with a typical sigmoidal curve. • Complex cysts may increase and decrease in volume spontaneously over time. • The fast growth rate of solid tumours correlates with their latest volume. • The fast growth rate can orient the scheduling of subsequent follow-ups. Keywords von Hippel-Lindau disease . Growth curve . Renal mass . Active surveillance . Magnetic resonance imaging Abbreviations CCs Complex cysts DT Doubling time FGR Fast growth rate FU Follow-up GR Growth rate MRI Magnetic resonance imaging RCC Renal cell carcinoma RMs Renal masses STs Solid tumours VHL Von Hippel-Lindau
Introduction von Hippel-Lindau (VHL) disease is a rare, autosomal dominant, inherited neoplastic disorder characterized by the development of a variety of benign and malignant tumours, including retinal and central nervous system hemangioblastomas, renal cysts and tumours, and pancreatic cysts and tumours,
Eur Radiol
among others [1, 2]. The prevalence of VHL disease is estimated at one in 36,000 births, and its penetrance is over 90 % by the age of 65 [3]. It is caused by a mutation of the VHL gene located on chromosome 3p25 [4]. Renal cell carcinoma (RCC) is one of the most common causes of death in patients with VHL disease and is seen in 24–45 % of patients. When renal cysts are also considered, 60 % of VHL patients have renal lesions. RCC often remains asymptomatic for long periods of time, and serial imaging is needed for its early diagnosis [1, 5]. Complex cysts (CCs) are fluid cystic masses, often containing solid RCC components, and are usually treated as potentially malignant lesions. They need to be monitored, although transitions from CCs to solid lesions are quite rare [1, 6]. Renal masses (RMs) in VHL patients are often multiple and bilateral, and their growth rate varies considerably, complicating their management, but the general strategy is to delay surgical treatment for as long as possible while keeping the risk of metastases reasonably low. Active surveillance is needed to enable kidney function for as long as possible, to avoid dialysis, and limit the risk of invasive disease, with serial cross-sectional imaging to monitor a lesion’s growth until its size (maximum diameter) reaches a threshold conventionally set at 3 cm. This is because it has been demonstrated that RCCs less than 3 cm in diameter rarely metastasize [1, 7, 8]. When STs reach or exceed this 3-cm threshold, several surgical treatments are possible, including radical nephrectomy, nephron-sparing surgery, and the more recently introduced minimally-invasive techniques (radiofrequency ablation and cryoablation) [9–11]. There are no specific guidelines on the timing of the FU for VHL patients, but MRI imaging is usually scheduled once a year. Experimental cancer research on animals has demonstrated, however, that tumours grow exponentially only as long as they are small, while their growth rate slows down as they become larger (as in the Gompertzian or logistic models) [12–14]. In clinical studies, estimates of the rate of volumetric growth of untreated tumours are usually based on brief intervals, for which the exponential model can be applied to describe a tumour’s growth, and endpoints such as doubling time (DT) can be used to assess a tumour’s aggressiveness [15]. As mentioned earlier, tumours in VHL patients are usually left to grow untreated until they reach a given size threshold [16, 17], and a non-exponential model of growth - with its typical sigmoidal shape - is assumed to be applicable. Actual tumour growth can be checked in vivo by means of an active surveillance program for VHL patients, measuring the volume of their lesions by means of imaging at each FU. The purpose of the study was to estimate an optimal FU interval between consecutive MRI examinations in VHL patients with RMs, based on the assessment of growth curves and the determination of tumour growth rates.
Materials and methods Study population The study was approved by the Institutional Review Board; the need to obtain patients’ informed consent was waived, given the retrospective nature of the study. Forty-four consecutive patients with genotype-confirmed VHL disease were identified using the institutional electronic medical record system to obtain demographic information, relevant medical history, pertinent radiology, and pathology reports. Of these 44 VHL patients, 26 (59.1 %) had renal involvement and comprise the study population. These patients all had RMs, either solid tumours (STs), or CCs, or both, in one or both kidneys. Twelve of them had previously undergone surgery for RCC, involving either nephrectomy, or nephron-sparing surgery. Simple cysts [18] were not considered in this study. All patients had been under active surveillance with a periodical MRI for a total period ranging from 7 to 104 months (mean: 50.3 months) with variable FU intervals. In vivo RM growth curves were constructed and analyzed by determining the volume of 63 RMs in all MRIs obtained during the surveillance period. Imaging technique Active surveillance for VHL patients was based on MRI. All imaging performed at our institute was done with a 1.5 T unit (Magnetom Espree; Siemens Medical Systems, Erlangen, Germany) with phased-array coils (four-element anterior and eight-element posterior coil arrays) to increase the signal-to-noise ratio. Each series of images included the following: (1) axial and coronal breath-hold T2-weighted MRI with half-Fourier singleshot turbo spin-echo sequences; (2) axial in- and opposed-phase T1-weighted spoiled gradient-echo sequences; (3) axial three-dimensional fat-suppressed T1weighted interpolated spoiled gradient-echo sequences (volumetric interpolated breath-hold examination, VIBE) acquired before and during arterial, venous, and delayed phases, after intravenous administration of contrast medium (0.2 mL/kg up to a maximum of 16 mL; flow rate 2–2.5 mL/sec; Omniscan; GE Healthcare A.S., Oslo, Norway). Parameter settings for each MRI sequence are summarized in Table 1. Some MRI examinations performed at other hospitals (31 out of 162; 19.1 %) were included in the study because sequences had been obtained before and after injection of the contrast medium, and the images were of adequate diagnostic quality.
Eur Radiol Table 1 Parameter settings for each sequence used in MRI examinations for VHL patients included in the study Parameter
T2-weighted
T1-weighted
VIBE
Repetition time (msec) Echo time (msec)
1,500 93
3.8 1.4
Flip angle (°) Matrix (pixel) Section thickness (mm) Intersection gap (mm) Field of view (mm2)
170 180×320 6 1.2 400×300
129 4.9 in phase 2.4 out phase 70 162×256 6 1.2 380×308
10 150×320 2.3 0.5 420×301
Image analysis All MRI images were examined independently by two experienced radiologists (F.P., C.D.B.), who assessed the characteristics of all RMs, classifying them as STs or CCs, and measured three maximum diameters (two axial and one coronal) of each lesion to calculate their volume. If the lesion was classified in the same way by both radiologists and the differences between their measurements were small (1 cm3/month), suggesting that the FGR could be used to adjust the FU interval between two consecutive imaging procedures in VHL patients with a view to closely monitoring ST growth. In fact, the FGR provides a quantitative indication of a lesion’s growth and could help clinicians to schedule subsequent MRI examinations; an optimal FU interval should be established depending on the FGR, the latest-recorded volume of a given mass, and the overall stage of the growth curve. Table 3 suggests the FU interval associated with a given FGR and latest volume, assuming that the growth curve or a portion of it is known.
4
6
8
10
12
14
16
18
20
Vollatest (cm3)
The first condition to meet in order to use these criteria is to obtain enough information about the growth curve to be able to ascertain the FGR. If a new, small lesion is discovered (volume
