Probands are marked with a black arrow. An orange diamond denotes individuals enrolled in the study, and filled blue symbols denote confirmed macular telangiectasia diagnoses.
Fundus photograph of a 21-year-old proband with MacTel (proband in Figure 2B, subfamily on the right) shows mild temporal pigment clumping (A) compared with more extensive pigment clumping in the proband’s father (C). Macular pigment scanning shows central loss of luteal pigment with redistribution into a circular ring of luteal pigment in both the proband (B) and his father (D)
Early and late fluorescein angiography in the 21-year-old proband with MacTel (A and B) and his father (C and D) from Figure 2B. There is early delineation of juxtafoveal telangiectatic vessels in both the proband (A) and his father (C) with intense perifoveal hyperfluorescence at later times (B and D).
eFigure 1. Blue light reflectance images and macular optical coherence tomography of two MacTel patients.
eFigure 2. J28/F289 Family
eFigure 3. UT001/F287 Family
eFigure 4. UT002/F288 Family
eFigure 5. O15/F160 Family
eFigure 6. M39/F175 Family
eFigure 7. K80/F213 Family
eFigure 8. D08/F243 Family
eFigure 9. UT014/F293 Family
eFigure 10. R44 Family
eFigure 11. UT007 Family
Customize your JAMA Network experience by selecting one or more topics from the list below.
Ronquillo CC, Wegner K, Calvo CM, Bernstein PS. Genetic Penetrance of Macular Telangiectasia Type 2. JAMA Ophthalmol. 2018;136(10):1158–1163. doi:10.1001/jamaophthalmol.2018.3283
What is the apparent genetic penetrance of macular telangiectasia type 2 (MacTel) under an autosomal dominant model?
In this cross-sectional study, a collection of large, well-phenotyped MacTel families showed an apparent genetic penetrance of 0.38 (95% CI, 0.19-0.57).
There is incomplete genetic penetrance of MacTel, and these families may be of value for the identification of MacTel genes.
The apparent genetic penetrance of macular telangiectasia type 2 (MacTel) is important for gene discovery studies and for clinical risk assessment of affected individuals’ family members.
To determine the genetic penetrance of MacTel.
Design, Setting, and Participants
Descriptive cross-sectional study of patients with MacTel at a tertiary referral eye center. From 2008 to 2016, consecutive patients with MacTel were independently identified, and all of their available siblings and parents were recruited. Seventeen probands with MacTel were included in the study who satisfied the requirement of having at least 1 parent or sibling willing and able to participate. Data from these 17 families were included for the analysis of apparent genetic penetrance.
Main Outcomes and Measures
Determination of MacTel genetic penetrance in probands’ parents and siblings.
Of 80 study participants, 50 (62.5%) were women. The mean (SD) age of study participants with MacTel was 61.2 (14.0) years (range, 23-81 years) and without MacTel was 60.7 (16.4) years (range, 24-92 years). There were 17 MacTel probands, and there was a high rate of enrollment of living siblings and parents: 52 of 71 living siblings (73%) and 11 of 12 parents (92%). Of 52 enrolled siblings, 9 (17%) were affected. Of 11 enrolled parents, 3 (27%) had MacTel. Apparent genetic penetrance was calculated to be 0.35 (95% CI, 0.14-0.6) by sibling analysis and 0.55 (95% CI, 0.02-1.00) by parent analysis. Combining the sibling and parent analyses, the apparent penetrance was calculated to be 0.38 (95% CI, 0.19-0.57).
Conclusions and Relevance
The genetic penetrance of MacTel in rigorously phenotyped multiple large families is described. Families such as these could be critical for successful identification of MacTel genes.
Macular telangiectasia type 2 (MacTel) was originally described by several researchers as an idiopathic juxtafoveal retinal telangiectasis.1-3 MacTel is characterized by pigment clumping and telangiectatic vessels temporal to the macula, commonly with outer and inner temporal juxtafoveal cavitations and thinning. Juxtafoveal telangiectatic vessels are highlighted during the early stages of fluorescein angiography with temporal or perifoveal leakage and/or staining at later timeframes, and there is loss or redistribution of central macular pigment.3-5 Increased confocal blue-light reflectance has also been demonstrated, but the mechanism of this phenomenon has not been elucidated.5 There are several reports showing vertical transmission of disease suggestive of an autosomal dominant mode of inheritance,6-8 but if this is the case, it is not fully penetrant because the siblings’ prevalence of MacTel in these pedigrees is far below 50%. Although multiple genetic loci have been associated with MacTel, causative genetic mutations have yet to be identified.9
To our knowledge, there are no published reports calculating a genetic penetrance estimate for MacTel. Accurately and precisely calculating genetic penetrance requires sufficiently large families with multiple affected individuals and rigorous phenotyping of as many family members as possible. Here we report a cross-sectional study of well-phenotyped MacTel family pedigrees from Utah and Idaho and calculate the apparent genetic penetrance of disease among the probands’ parents and siblings.
This study was approved by the University of Utah institutional review board and complied with the Declaration of Helsinki.10 Patients with MacTel were identified from clinical visits and records at the Moran Eye Center, as well as by referrals from local and regional ophthalmologists in the Intermountain West. Written informed consent was obtained prior to enrollment of study participants, and we complied with all requirements per the Health Insurance Portability and Accountability Act of 1996. Clinical phenotyping included a baseline eye examination, optical coherence tomography of the macula, fluorescein angiography, fundus autofluorescence, blue-light reflectance, and measurement of macular pigment density and distribution by dual-wavelength autofluorescence imaging. A single retina specialist (P.S.B.) examined all participants and determined affected or nonaffected status based on the combination of clinical, angiographic, and optical coherence tomography findings. The images were then sent to the Moorfields Eye Hospital MacTel Reading Centre in London for masked confirmation of the clinician’s diagnosis. Family pedigrees were constructed, and probands with living siblings and parents in the region were selected for further familial investigations. All available siblings and parents were then asked to participate in the study and received the same testing. When a proband’s parent was identified as affected, that parent’s siblings were also recruited for the study.
We used SPECTRALIS (Heidelberg Engineering) set on IR (for infrared images), FA (for autofluorescent images and fluorescein angiography), and red-free (for blue-light reflectance images) to take 30° fields with the fovea centered in both eyes. Optical coherence tomography was obtained on the SPECTRALIS with the 30° crosshair centered on the fovea, imaging 97 sections at 30-μm intervals. Macular pigment images were taken in alternating blue autofluorescence and green autofluorescence in a 30-second movie using special software (Heidelberg Engineering). The images were then calculated and graphed as a measure of pigment density measured from the foveal center moving outward concentrically.
The Utah Population Database contains millions of genealogical records and demographic data of residents of the Intermountain West. Demographic information to our MacTel probands was submitted to the Utah Population Database to ascertain previously unrecognized familial relationships. From this query, 2 MacTel superfamilies were identified. Three family pedigrees were combined to make the W62/F204 superfamily, and 2 family pedigrees were combined to make the M65/F172 superfamily.
Seventeen probands enrolled between 2008 and 2016 met the study criteria of having at least 1 living sibling or parent willing and able to be examined for this study. The Table shows the demographic information of study participants. All study participants were of European heritage except for 1 family of Persian descent. A total of 71 living siblings and 12 living parents (4 parental couples) were identified from the pedigrees during enrollment. High enrollment was achieved with 52 of 71 siblings (73%) and 11 of 12 parents (92%) enrolled in the study. Of 52 examined siblings, 9 (17%) were found to have MacTel (Table). Seven independently identified families had 2 to 4 affected first-degree family members including the proband (Figure 1 and Figure 2). Of 11 examined parents, 3 (27%) had MacTel. There was an affected parent in 3 of 4 parental couples.
The mean (SD) age of patients with MacTel was 61.2 (14.0) years compared with the mean (SD) age of unaffected parents and siblings of 60.7 (16.4) years (95% CI, −36.31 to 49.91; P = .88). The enrolled participants were predominantly women (50 [62.5%]). Among all parents and siblings of probands, 5 of 25 men (20%) had MacTel and 7 of 38 women (18%) had MacTel. There was no significant sex difference between affected and unaffected participants (95% CI for odds ratio, 0.2121-4.4168).
Penetrance estimates for disease were calculated in 2 ways, by sibling analysis and by parent analysis using an autosomal dominant model. If p is the penetrance of a candidate gene causing MacTel, the probability that a sibling or parent is affected was calculated to be p/2 (ie, a fully penetrant autosomal dominant gene [p = 1.00] implies a 50% probability that a sibling, parent, or offspring would be affected). From sibling data, we estimate the penetrance to be p = 2 × (9/52) or p = 0.35 (95% CI, 0.14-0.6). Calculating the penetrance by analyzing the parent data, penetrance was found to be p = 2 × (3/11) or p = .55 (95% CI, 0.02-1.00). Combining siblings and parents, we estimate the penetrance at 2 × 12/63 = 0.38 (95% CI, 0.19-0.57).
One of the earliest families that we identified is shown in Figure 1A (P48/F156 family; this is an updated and expanded pedigree from Figure 1B in the study by Parmalee et al8). The proband, her 6 living siblings, and both parents were all enrolled in the study. Two of 6 siblings were found to have MacTel, although they were clinically asymptomatic. Several offspring of the proband and of 1 affected sibling were also examined. None of the offspring showed signs of the disease. The proband’s father was unaffected, while the mother’s previous diagnosis of MacTel was confirmed. The proband’s affected mother had 4 siblings. Two were enrolled in the study, and both were found to be unaffected. Furthermore, 5 first cousins of the proband were enrolled in the study because of complaints of visual impairment, but all were found to be normal.
The Z10/F276 family pedigree is shown in Figure 1B. The proband has 4 living siblings (all female) and 2 living parents. Three of 4 siblings and both parents were enrolled in the study. We found that 1 sibling and the proband’s father were affected, although both were asymptomatic. The proband’s father had 1 living sibling who was unaffected. The proband’s maternal first cousin was also enrolled in the study because of reports of visual loss but was not affected by MacTel.
Figure 1C shows the pedigree of the F33/F290 family. A local ophthalmologist diagnosed a pair of siblings in this family as having MacTel and referred them to our clinic. The siblings, 1 male and 1 female, were confirmed to have MacTel. They have 4 other siblings who are all unaffected. Both parents had died before the study. The proband’s maternal aunt was reported to have MacTel but was found to be normal.
In the P39/UT033 family (Figure 1D), we diagnosed 1 of the proband’s brothers as having MacTel that was previously unrecognized. She had 5 other sisters who are all unaffected. Both of her parents were enrolled in the study, but neither the mother nor the father had MacTel. The proband’s paternal grandmother was also unaffected.
All of the MacTel families identified in this study were from Utah or Idaho. None of the families had known consanguinity; however, given the geographic constraints of our population, we hypothesized that some of the probands might share a common ancestor. We submitted demographic data on all study participants with MacTel to the Utah Population Database to find previously unrecognized family relationships. We call the pedigrees that we assembled MacTel superfamilies. Figure 2A shows the pedigree for the W62/F204 superfamily (this is an updated and expanded pedigree from Figure 1D in the study by Parmalee et al8). Three subfamilies collectively make up this superfamily. The proband in the middle subfamily had 8 of 9 siblings still alive. Six were enrolled in the study, and 3 of her sisters were found to have MacTel. None of the siblings of the probands in the left subfamily and in the right subfamily were available to participate in the study. The parents of the probands in the subfamilies to the left and center had both died. The mother of the proband in the right subfamily was still alive but was not able to participate in the study.
The M65/F172 superfamily is our second superfamily composed of 2 subfamilies that were not previously known to be related (Figure 2B). The proband in the left side of the pedigree had 1 of her 5 siblings affected. Both parents had died. The proband in the right side of the pedigree had 2 sisters. Only 1 had been enrolled in the study, and she was found to be unaffected. Both of the proband’s parents were still alive and were enrolled in the study. His mother was found to be normal. His father had been previously diagnosed as having macular degeneration, but he actually had MacTel. The proband’s affected father also had 2 living brothers. One of the brothers was enrolled in the study and was found to be unaffected.
The proband in the right side of the M65/F172 superfamily, originally referred to our clinic with a diagnosis of cone dystrophy, is one of the youngest individuals ever diagnosed as having MacTel at age 21 years, to our knowledge. Figure 3 and Figure 4 show the clinical phenotype of this patient alongside his 52-year-old affected father. Both share characteristic MacTel findings including pigment clumping in the macula (Figure 3A and C), decreased central macular pigment density and redistribution in a circular pattern at 5° to 6° of foveal eccentricity (Figure 3B and D), juxtafoveal telangiectatic vessels (Figure 4A and C) with intense hyperfluorescence at later times (Figure 4B and D), increased parafoveal ring-like reflectivity with blue light reflectance (eFigure 1A and C in the Supplement), and temporal ellipsoid zone loss (eFigure 1B and D in the Supplement). The pedigrees of the remaining 10 families who had no affected siblings or parents are provided in eFigures 2 to 11 in the Supplement.
Macular telangiectasia type 2 is an idiopathic macular disease with characteristic clinical and imaging findings.4-6 It is widely accepted that MacTel can be a familial disease, and recent studies suggest that MacTel is dominantly inherited with incomplete penetrance.8 As part of the MacTel Project, we prospectively enrolled patients to this study and recruited all available living siblings and parents to determine the apparent genetic penetrance.
A high rate of enrollment of living siblings (73%) and parents (92%) of probands was achieved in this study. There was limited bias in recruitment, as we made an effort to recruit all available siblings and parents of the affected individuals whether or not they had a history of vision impairment. The Moorfields Eye Hospital MacTel Reading Centre also independently reviewed all of the individuals’ clinical imaging results in a masked fashion and confirmed the affected and unaffected status with 100% diagnostic concordance with the clinician.
Our results show vertical transmission of disease from male to female and female to male consistent with dominant inheritance (Figure 1A and B and Figure 2B). Our results are also consistent with incomplete penetrance given the finding that we did not find an affected parent in family P39/UT003 (Figure 1D). The apparent genetic penetrance was calculated to be 0.35 (95% CI, 0.14-0.6) by sibling analysis and 0.55 (95% CI, 0.02-1.00) by parent analysis. Given the smaller number of individuals available for parental analysis, genetic penetrance is likely closer to the penetrance calculated by sibling analysis. A combined analysis gives an apparent genetic penetrance estimate of 0.38 (95% CI, 0.19-0.57).
Although we had a very high recruitment of siblings for this study, it is possible that the calculation of genetic penetrance is confounded by incomplete recruitment because affected siblings may be more likely to participate than unaffected siblings. However, from our experience, only a few of the affected first-degree relatives had a prior diagnosis of MacTel, and some relatives with a prior diagnosis of MacTel or nonspecific vision loss turned out not to have MacTel. Nevertheless, if we were to assume that all of the siblings who did not participate in the study did not have MacTel, our apparent genetic penetrance decreases to 0.25 (95% CI, 0.1-0.41) by sibling analysis (9 affected siblings of 71 total siblings).
Previous studies have shown that MacTel can remain clinically asymptomatic.6 Therefore, a normal clinical history is not sufficient to exclude MacTel. As we have made the effort to recruit all available siblings and parents, we have made new diagnoses of MacTel in several of the probands’ siblings and parents. Our experience agrees with previous studies that MacTel can remain asymptomatic with good visual acuity.6 In our study, we found no significant difference between the age of affected individuals and unaffected parents and siblings, implying that relatively few of the clinically unaffected siblings and parents are likely to convert to affected status in the future. Indeed, a number of unaffected individuals enrolled in the early years of the study were reexamined about 5 years later, and no new MacTel diagnoses were made.
The offspring of affected individuals are typically excluded from MacTel genetic studies because being 20 to 40 years younger than their affected parents, they are unlikely to manifest a disease that rarely is diagnosed before age 40 years. In this study, we diagnosed a 21-year-old individual as having MacTel (Figure 2B, Figure 3, and Figure 4), one of the youngest patients ever reported to have the disease, to our knowledge.11 When we examined some adult children of affected siblings and probands, none were affected (Figure 1A and Figure 2B). It is likely that some of the children will develop MacTel in the future but currently have no manifestations of the disease given their young age. As there are no genetic screening tests available, it may be possible for new imaging modalities to detect subtle changes in early MacTel. A recent report shows that fluorescence lifetime imaging ophthalmoscopy is a novel imaging modality to detect characteristic retinal changes in patients with MacTel.12 In the P39/UT033 family (Figure 1D), both clinically unaffected parents of the proband were imaged with fluorescence lifetime imaging ophthalmoscopy, and we detected retinal abnormalities suggestive of MacTel only in the mother.12 It will be interesting to follow this individual to determine if she eventually develops clinical findings of MacTel to further validate fluorescence lifetime imaging ophthalmoscopy as an early imaging modality for MacTel.
We have assumed that MacTel is an autosomal dominant disease, but it is possible that there are multiple different dominant gene mutations that have different genetic penetrance, and therefore, our results could be an apparent or blended penetrance estimate. The Utah Population Database is a powerful and unique resource to create MacTel superfamilies, but the fact that these may be rather distant relationships warrants caution that distinct genes could be causative in the subfamilies. It is possible, but less likely, that MacTel is recessively inherited, as suggested by MacTel superfamily W62/F204, which has affected maternal and paternal cousins of the central family. On the other hand, our 3 families with vertical transmission from parent to child argue against recessive inheritance.
The major limitation of this study is the lack of genotypic information to correlate with the clinical phenotype. Although there is no causative MacTel gene mutation that has been identified thus far, 3 genetic loci associated with MacTel have now been identified by genome-wide association studies.9 It will be interesting to correlate these risk loci to these families.
We predict that large and well-phenotyped MacTel families, including those described in this article, will be important for identification of MacTel genes, but our calculated apparent genetic penetrance of 38% and the potential for multiple causative genes highlight the challenges ahead for MacTel gene discovery. It will be important to continue to recruit large families with multiple affected individuals until we understand the genetic basis of this disease.
Accepted for Publication: May 25, 2018.
Corresponding Author: Paul S. Bernstein, MD, PhD, John A. Moran Eye Center, University of Utah, 65 Mario Capecchi Dr, Salt Lake City, UT 84132 (email@example.com).
Published Online: August 9, 2018. doi:10.1001/jamaophthalmol.2018.3283
Author Contributions: Drs Ronquillo and Bernstein had full access to all of the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis.
Concept and design: Bernstein.
Acquisition, analysis, or interpretation of data: All authors.
Drafting of the manuscript: Ronquillo, Calvo, Bernstein.
Critical revision of the manuscript for important intellectual content: All authors.
Statistical analysis: Ronquillo.
Obtained funding: Bernstein.
Administrative, technical, or material support: Wegner, Calvo.
Conflict of Interest Disclosures: All authors have completed and submitted the ICMJE Form for Disclosure of Potential Conflicts of Interest and none were reported.
Funding/Support: This study was funded by the Lowy Medical Research Institute of Sydney, Australia. Departmental support was provided by an unrestricted grant from Research to Prevent Blindness and a National Institutes of Health core grant (EY-14800). Dr Ronquillo is a recipient of an Achievement Rewards for College Scientists Scholar grant.
Role of the Funder/Sponsor: Lowy Medical Research Institute funded this study and facilitated independent masked grading of the fundus images by Moorfields Eye Hospital MacTel Reading Centre. They had no additional role.
Meeting Presentation: Some of the data included in this paper were presented at The Association for Research in Vision and Ophthalmology Annual Meeting; May 7, 2017; Baltimore, Maryland and at the Macula Society 40th Annual Meeting; June 9, 2017; Singapore.
Additional Contributions: We thank Kelliann Farnsworth, COT; Kara Halsey, BS; Briana Sawyer, MS; and Barbara Hart, BS (all from John A. Moran Eye Center), for coordinating study visits and constructing the family pedigrees. We also thank the referring physicians from all throughout the Intermountain West. We thank Alun Thomas, PhD (Genetic Epidemiology, University of Utah), for providing statistical calculations and James Gilman, COT (John A. Moran Eye Center), for preparing the figures. No compensation was received.