Quarterly Journal of Medicine, New Series 77, No. 283, pp. 1151-1163, November 1990
Clinical Features and Natural History of von Hippel-Lindau Disease E. R. MAHER, J. R. W. YATES, R. HARRIES*, C. BENJAMINf, R. HARRISf, A. T. MOORE}, and M. A. FERGUSON-SMITH Cambridge University Department of Pathology, Tennis Court Road, Cambridge; *Department of Radiology, Grimsby District General Hospital; |Department of Medical Genetics, University of Manchester; \Department of Ophthalmology, Addenbrooke's Hospital, Cambridge, England. Accepted 3 April 1990
The clinical features, age at onset and survival of 152 patients with von Hippel-Lindau disease were studied. Mean age at onset was 26.3 years and 97 per cent of patients had presented by aged 60 years. Retinal angioma was thefirstmanifestation in 65 patients (43 per cent), followed by cerebellar haemangioblastoma (n = 60,39 per cent) and renal cell carcinoma (n = 15,10 per cent). Overall, 89 patients (59 per cent) developed a cerebellar haemangioblastoma, 89 (59 per cent) a retinal angioma, 43 (28 per cent) renal cell carcinoma, 20 (13 per cent) spinal haemangioblastoma and 11 (7 per cent) a phaeochromocytoma. Renal, pancreatic and epididymal cysts were frequent findings but their exact incidence was not accurately assessed. Mean age at diagnosis of renal cell carcinoma (44.0 ± 10.9 years) was significantly older than that for cerebellar haemangioblastoma (29.0 ±10.0 years) and retinal angioma (25.4 ±12.7 years). The probability of a patient with von Hippel-Lindan disease developing a cerebellar haemangioblastoma, retinal angioma or renal cell carcinoma by age 60 years was 0.84,0.7 and 0.69, respectively. A comprehensive screening protocol for affected patients and at-risk relatives is presented, based on detailed analysis of age at onset data for each of the major complications. Median actuarial survival was 49 years, with renal cell carcinoma the leading cause of death. INTRODUCTION Von Hippel-Lindau disease is an autosomal dominant inherited disorder characterized by a predisposition to develop a wide variety of tumours, most frequent haemangioblastomas of the central nervous system and retina, renal cell carcinoma, phaeochromocytoma and renal, pancreatic and epididymal cysts [1-3]. The gene for von Hippel-Lindau disease have been mapped to the short arm of chromosome 3, but localization is not yet precise enough for presymptomatic or prenatal diagnosis. Address correspondence to: Dr E. R. Maher, Department of Medical Genetics, Level I, Addenbrooke's Hospital, Hills Road, Cambridge CB2 2QQ, UK. ©Oxford University Press 1990
Downloaded from by guest on May 23, 2015
SUMMARY
1152
E. R. Maher and others
The history of von Hippel-Lindau disease has been described [1, 3]. Following early descriptions of retinal angiomatosis by Collins [4] and von Hippel [5,6], Lindau [7] described the retinal, central nervous system and visceral manifestations of this pleiotropic disorder. Melmon and Rosen [1] proposed the widely accepted criteria for the diagnosis of von HippelLindau disease: two or more haemangioblastomas, or a single haemangioblastoma in association with a visceral manifestation. Where there is a family history of retinal or central nervous system haemangioblastoma, only one haemangioblastoma or visceral complication is required. By 1987 more than 500 cases of von Hippel-Linden disease had been described. Most previous studies have consisted of small series of patients presenting to individual specialties, or reports of single large families with many affected members [8,9]. Both of these methods of ascertainment may produce bias. We report the largest series of previously unreported patients recruited from various medical specialties throughout Great Britain. METHODS
RESULTS First Manifestation of von Hippel-Lindau Disease The most frequent initial manifestation of the disease was retinal angiomatosis (n = 65) followed by cerebellar haemangioblastoma (n = 60), renal cell carcinoma (n= 15), phaeochromocytoma (H = 8) and spinal haemangioblastoma (n = 4). Twenty-two of the 152 patients were identified through screening and were asymptomatic at presentation (19 had retinal angiomatosis, two had renal cell carcinoma and 1 had phaeochromocytoma). An age at first diagnosis curve was constructed for all 152 patients (Fig. 1). Mean age at first diagnosis of von Hippel-Lindau disease was 26.3 years (median = 23.5 years). The proportion of all patients who had presented by 30,40, 50 and 60 years were 66, 86, 95, and 97 per cent, respectively. Exclusion of patients detected by asymptomatic screening had minimal effect on the age at diagnosis curve; the proportions of patients who were symptomatic at first diagnosis who had presented by 20,40, 50 and 60 years were 65, 85, 95, and 98 per cent, respectively. M e a n ± S D age at presentation in symptomatic cases was
Downloaded from by guest on May 23, 2015
Patients with von Hippel-Lindau disease were ascertained for the purpose of molecular genetic studies by contacting specialists in Medical Genetics, Nephrology, Neurology, Neurosurgery, Ophthalmology and Urology throughout Great Britain. Details of both sporadic and familial cases were requested. Clinical details were obtained through patient interview, hospital notes and autopsy reports of all the affected patients referred and for other affected family members who were then identified. Particular attention was paid to the incidence, age at diagnosis and mode of detection for the five lesions which cause most mortality and morbidity: cerebellar and spinal haemangioblastoma, retinal angiomas, renal cell carcinoma and phaeochromocytoma. The basis of this report are 152 affected patients (79 males and 73 females, 101 alive and 51 dead; 12 isolated cases and 140 familial from 45 kindreds) on whom sufficient clinical details were available and who had not previously been reported. All patients satisfied the diagnostic criteria for von Hippel-Lindau disease defined by Melmon and Rosen [1] (see above). Genetic aspects of the disease will be reported in detail elsewhere. Statistical analysis was performed using unpaired /-testing and the x2 test with Yates', correction as appropriate. Cumulative risk and actuarial survival were calculated using life table methods [10]. Statistical significance was taken at the 5 per cent level.
von Hippel-Lindau
Disease
1153
1.00.9-
10 n c
ort
o n
P
0.70.60.5-
040.30.2-
0 1-
on10
—r— 20
30
—I—
—I—
40
50
—r— 60
—I
70
Age (years) FIG. 1. Cumulative age at onset curve in 152 patients with von Hippel-Lindau disease.
Complications of von Hippel-Lindau Disease The incidence of the major complication of disease found in this study was compared to that in 554 patients reported in the literature (Table 1). Seventy-four of 152 patients developed a major complication (retinal angioma, central nervous system haemangioblastoma, renal cell carcinoma, phaeochromocytoma) at two or more sites: 50 had retinal angioma and central nervous system haemangioblastoma, 19 had retinal angioma and renal cell carcinoma, and 21 had a central nervous system haemangioblastoma and renal cell carcinoma. Eleven
Downloaded from by guest on May 23, 2015
27.0 ± 12.6 years, and 22.7 ± 14.4 years in asymptomatic patients (/ = 1.45, p = 0.15). There was no difference in mean age at presentation in males and females (25.1 ±12.9 and 27.7± 12.9, respectively, r = 1.24, /? = 0.22). In familial cases the age at onset was not influenced by whether the affected parent was the father (mean±SD 23.4 ±11-8 years, n = 57) or mother (24.4± 10.5 years, n = 43). Mean age at diagnosis of von Hippel-Lindau disease in isolated cases was not significantly different from familial cases (22.3 ±7.3 vs. 26.7 ± 13.4, t= 1.12 p = 0.26). Mean age±SD (range) for patients presenting with retinal angiomatosis was 21.8 ± 11.5 (4-68) years, for patients presenting with cerebellar haemangioma, 27.8±9.9 (13-61) years, and for renal cell carcinoma 46.5± 11.5 (24-67) years. For familial cases probands (n = 45) were compared to secondary cases (n = 95): there were no significant differences between the two groups in mean age at first diagnosis or current age (probands 25±8.7 and 36.1 ±10.3 years, respectively; secondary cases 27.6± 14.8 and 35 ± 15.7 years, respectively). However 19 of 95 secondary cases had been diagnosed by screening of asymptomatic individuals compared to three of 45 probands (^2 = 3.1, 0.05
Clinical Features and Natural History of von Hippel-Lindau Disease E. R. MAHER, J. R. W. YATES, R. HARRIES*, C. BENJAMINf, R. HARRISf, A. T. MOORE}, and M. A. FERGUSON-SMITH Cambridge University Department of Pathology, Tennis Court Road, Cambridge; *Department of Radiology, Grimsby District General Hospital; |Department of Medical Genetics, University of Manchester; \Department of Ophthalmology, Addenbrooke's Hospital, Cambridge, England. Accepted 3 April 1990
The clinical features, age at onset and survival of 152 patients with von Hippel-Lindau disease were studied. Mean age at onset was 26.3 years and 97 per cent of patients had presented by aged 60 years. Retinal angioma was thefirstmanifestation in 65 patients (43 per cent), followed by cerebellar haemangioblastoma (n = 60,39 per cent) and renal cell carcinoma (n = 15,10 per cent). Overall, 89 patients (59 per cent) developed a cerebellar haemangioblastoma, 89 (59 per cent) a retinal angioma, 43 (28 per cent) renal cell carcinoma, 20 (13 per cent) spinal haemangioblastoma and 11 (7 per cent) a phaeochromocytoma. Renal, pancreatic and epididymal cysts were frequent findings but their exact incidence was not accurately assessed. Mean age at diagnosis of renal cell carcinoma (44.0 ± 10.9 years) was significantly older than that for cerebellar haemangioblastoma (29.0 ±10.0 years) and retinal angioma (25.4 ±12.7 years). The probability of a patient with von Hippel-Lindan disease developing a cerebellar haemangioblastoma, retinal angioma or renal cell carcinoma by age 60 years was 0.84,0.7 and 0.69, respectively. A comprehensive screening protocol for affected patients and at-risk relatives is presented, based on detailed analysis of age at onset data for each of the major complications. Median actuarial survival was 49 years, with renal cell carcinoma the leading cause of death. INTRODUCTION Von Hippel-Lindau disease is an autosomal dominant inherited disorder characterized by a predisposition to develop a wide variety of tumours, most frequent haemangioblastomas of the central nervous system and retina, renal cell carcinoma, phaeochromocytoma and renal, pancreatic and epididymal cysts [1-3]. The gene for von Hippel-Lindau disease have been mapped to the short arm of chromosome 3, but localization is not yet precise enough for presymptomatic or prenatal diagnosis. Address correspondence to: Dr E. R. Maher, Department of Medical Genetics, Level I, Addenbrooke's Hospital, Hills Road, Cambridge CB2 2QQ, UK. ©Oxford University Press 1990
Downloaded from by guest on May 23, 2015
SUMMARY
1152
E. R. Maher and others
The history of von Hippel-Lindau disease has been described [1, 3]. Following early descriptions of retinal angiomatosis by Collins [4] and von Hippel [5,6], Lindau [7] described the retinal, central nervous system and visceral manifestations of this pleiotropic disorder. Melmon and Rosen [1] proposed the widely accepted criteria for the diagnosis of von HippelLindau disease: two or more haemangioblastomas, or a single haemangioblastoma in association with a visceral manifestation. Where there is a family history of retinal or central nervous system haemangioblastoma, only one haemangioblastoma or visceral complication is required. By 1987 more than 500 cases of von Hippel-Linden disease had been described. Most previous studies have consisted of small series of patients presenting to individual specialties, or reports of single large families with many affected members [8,9]. Both of these methods of ascertainment may produce bias. We report the largest series of previously unreported patients recruited from various medical specialties throughout Great Britain. METHODS
RESULTS First Manifestation of von Hippel-Lindau Disease The most frequent initial manifestation of the disease was retinal angiomatosis (n = 65) followed by cerebellar haemangioblastoma (n = 60), renal cell carcinoma (n= 15), phaeochromocytoma (H = 8) and spinal haemangioblastoma (n = 4). Twenty-two of the 152 patients were identified through screening and were asymptomatic at presentation (19 had retinal angiomatosis, two had renal cell carcinoma and 1 had phaeochromocytoma). An age at first diagnosis curve was constructed for all 152 patients (Fig. 1). Mean age at first diagnosis of von Hippel-Lindau disease was 26.3 years (median = 23.5 years). The proportion of all patients who had presented by 30,40, 50 and 60 years were 66, 86, 95, and 97 per cent, respectively. Exclusion of patients detected by asymptomatic screening had minimal effect on the age at diagnosis curve; the proportions of patients who were symptomatic at first diagnosis who had presented by 20,40, 50 and 60 years were 65, 85, 95, and 98 per cent, respectively. M e a n ± S D age at presentation in symptomatic cases was
Downloaded from by guest on May 23, 2015
Patients with von Hippel-Lindau disease were ascertained for the purpose of molecular genetic studies by contacting specialists in Medical Genetics, Nephrology, Neurology, Neurosurgery, Ophthalmology and Urology throughout Great Britain. Details of both sporadic and familial cases were requested. Clinical details were obtained through patient interview, hospital notes and autopsy reports of all the affected patients referred and for other affected family members who were then identified. Particular attention was paid to the incidence, age at diagnosis and mode of detection for the five lesions which cause most mortality and morbidity: cerebellar and spinal haemangioblastoma, retinal angiomas, renal cell carcinoma and phaeochromocytoma. The basis of this report are 152 affected patients (79 males and 73 females, 101 alive and 51 dead; 12 isolated cases and 140 familial from 45 kindreds) on whom sufficient clinical details were available and who had not previously been reported. All patients satisfied the diagnostic criteria for von Hippel-Lindau disease defined by Melmon and Rosen [1] (see above). Genetic aspects of the disease will be reported in detail elsewhere. Statistical analysis was performed using unpaired /-testing and the x2 test with Yates', correction as appropriate. Cumulative risk and actuarial survival were calculated using life table methods [10]. Statistical significance was taken at the 5 per cent level.
von Hippel-Lindau
Disease
1153
1.00.9-
10 n c
ort
o n
P
0.70.60.5-
040.30.2-
0 1-
on10
—r— 20
30
—I—
—I—
40
50
—r— 60
—I
70
Age (years) FIG. 1. Cumulative age at onset curve in 152 patients with von Hippel-Lindau disease.
Complications of von Hippel-Lindau Disease The incidence of the major complication of disease found in this study was compared to that in 554 patients reported in the literature (Table 1). Seventy-four of 152 patients developed a major complication (retinal angioma, central nervous system haemangioblastoma, renal cell carcinoma, phaeochromocytoma) at two or more sites: 50 had retinal angioma and central nervous system haemangioblastoma, 19 had retinal angioma and renal cell carcinoma, and 21 had a central nervous system haemangioblastoma and renal cell carcinoma. Eleven
Downloaded from by guest on May 23, 2015
27.0 ± 12.6 years, and 22.7 ± 14.4 years in asymptomatic patients (/ = 1.45, p = 0.15). There was no difference in mean age at presentation in males and females (25.1 ±12.9 and 27.7± 12.9, respectively, r = 1.24, /? = 0.22). In familial cases the age at onset was not influenced by whether the affected parent was the father (mean±SD 23.4 ±11-8 years, n = 57) or mother (24.4± 10.5 years, n = 43). Mean age at diagnosis of von Hippel-Lindau disease in isolated cases was not significantly different from familial cases (22.3 ±7.3 vs. 26.7 ± 13.4, t= 1.12 p = 0.26). Mean age±SD (range) for patients presenting with retinal angiomatosis was 21.8 ± 11.5 (4-68) years, for patients presenting with cerebellar haemangioma, 27.8±9.9 (13-61) years, and for renal cell carcinoma 46.5± 11.5 (24-67) years. For familial cases probands (n = 45) were compared to secondary cases (n = 95): there were no significant differences between the two groups in mean age at first diagnosis or current age (probands 25±8.7 and 36.1 ±10.3 years, respectively; secondary cases 27.6± 14.8 and 35 ± 15.7 years, respectively). However 19 of 95 secondary cases had been diagnosed by screening of asymptomatic individuals compared to three of 45 probands (^2 = 3.1, 0.05
