Francois C. Delori, Ph.D.

Harvard Medical School

Professor of Ophthalmology

Schepens Eye Research Institute of Massachusetts Eye and Ear

Senior Scientist

Research Summary

Center/Research Area Affiliations


Dr. Delori primarily studies the retinal pigmented epithelium (RPE) lipofuscin in aging, age-related macular degeneration (AMD), and Stargardt disease. As a biophysicist, he has developed a new technique to quantify retinal autofluorescence from images acquired with scanning laser ophthalmoscopes. This modality allows him to study the distribution of lipofuscin at and around pathological tissue (drusen, geographic atrophy, and hyperpigmentation), follow the progression of disease, and monitor the effect of treatment designed to decrease lipofuscin accumulation. Dr. Delori also has introduced a novel method to measure the optical density of macular pigment and is currently working on methods to measure the light transmission of the ocular media.


Ph.D., Imperial College of Science and Technology, London (1972)

His Story

Novel Retina Imaging Techniques

Dr. Delori’s field of expertise is the examination of the retina by noninvasive means. By shining various colors of light onto the retina and analyzing the nature of the light reflected by the retina, one can obtain quantitative information about many important biological parameters—such as oxygen levels in retinal blood vessels, speed of blood flow, diffusion of nutrients, and quantities of various pigments. (Pigments are colored substances that can be studied because they absorb light.)

Dr. Delori has pioneered novel imaging techniques of the retina and has developed advanced optical techniques to study the role of lipofuscin and melanin pigments in the retinal pigmented epithelium as well as new ways to measure the distribution of macular pigment in the neural retina. These techniques are now used worldwide for research in the field of aging and retinal degenerations, as well as for tools to diagnose disease.

Lipofuscin, melanin, and macular pigment are three pigments that may play important roles in the causation of AMD. The ability to measure these pigments may not only provide new basic information on AMD and other retinal degenerations, but may also allow novel diagnostic tests to detect early disease and monitor the effects of drug intervention.

A Look at the Center of Vision: the Macula

The macula is the region of the eye devoted to precise vision. Its center, the fovea, is about as large as the head of a pin (0.3 mm across) and contains about 115,000 cone photoreceptors cells, allowing for the detection of color and fine details. The image of two or three letters of this text on your retina occupies the entire fovea. Upon absorption of light, the photopigment (yet another type of pigment, typified by rhodopsin) in the photoreceptor cells undergoes chemical changes that produce electrical signals. These signals are conducted to the brain, where the image is interpreted. This activity is very demanding in energy and nutrients—photoreceptor cells have the highest metabolic rate of cells in the body—and is possible only through an exquisitely organized support and protection system.

The layer of retinal pigmented epithelium (RPE) cells is the rear lining of the retina, and is the main player in sustaining the health of the retinal photoreceptors. These cells transport nutrients for the photoreceptors from the choroid (a vascular layer that acts as source of oxygen for the deeper layers of the retina), and continuously renew the photopigment. As its name implies, the RPE also contains melanin, a very dark pigment. This pigment absorbs light reflected by the deeper retinal layers, thereby reducing unwanted light to the photoreceptors; melanin also protects the RPE by absorbing radiation and trapping noxious chemicals.

The photoreceptor/RPE system is additionally protected by the yellow macular pigment (composed of a mixture of carotenoids—lutein and zeaxanthin). This pigment is distributed in the cones of the retina. It absorbs blue light and thereby may help reduce photochemical damage in the photoreceptors and RPE. Furthermore, macular pigment has marked anti-oxidant properties, further reducing the risk of such damage. Macular pigment is not synthetized in the body but is acquired exclusively from the diet (spinach, broccoli, corn, and eggs are good dietary sources). Low amounts of macular pigment could result in increased photochemical damage to the photoreceptors and the RPE. Studies have shown that a low amount of carotenoids in the blood is related to increased severity of AMD. Thus, appropriate diets or supplementation is highly indicated for patients with AMD. Non-invasive methods to monitor changes in macular pigment in the retina will be critical to monitor the sucess of carotenoid therapy.

A technique to quantify this pigment was introduced by Dr. Delori, and has been adopted by many researchers. In recent experimental work in his own laboratory, Dr. Delori has shown that the distribution of macular pigment is highly variable among individuals and that its pattern of distribution depends upon age and gender.

Note that the RPE is separated from the choroid a by a thin layer of collagenous connective tissue called Bruch's membrane, which acts as a mechanical support for the RPE and photoreceptors. Because oxygen and nutrients diffuse across this membrane, its integrity is critical for the well-being of the photoreceptors and the RPE.

The high metabolic activity of the photoreceptors does not occur without damage. Their environment is rich in oxygen and light, causing continuous photochemical damage to the receptors. Therefore, photoreceptors are continuously renewed; the outer part of the receptor is chopped off and digested in the RPE. The resulting waste products diffuse through Bruch's membrane to the choroidal blood vessels. However, this process is not perfect; a minute fraction of the discarded photoreceptor material remains trapped in the RPE, where is accumulates as lipofuscin pigment (also called the “age-pigment”). This process is continuous (each photoreceptor is completely renewed in about ten days), and the lipofuscin accumulates throughout life. In old age, it may occupy a large part of the RPE cell and may compromise the normal activity of the cell.

Lipofuscin can be measured by means of its fluorescence. Dr. Delori introduced noninvasive techniques to quantify lipofuscin in the living eye. Studies using these techniques have shown that the amount of lipofuscin in the RPE is highly variable in old eyes, and that the rate of accumulation of lipofuscin is affected by genetic and environmental factors, such as smoking. In AMD, the levels of lipofuscin are highly variable and studies are currently aimed at understanding the progression of the pathology. Efforts by other laboratories are currently aimed at developing drugs that would decrease the accumulation of lipofuscin in the RPE. Quantification of lipofuscin by non-invasive means will be critical to assess the efficiency of these drugs.

Age-Related Macular Degeneration

The cause of AMD is not known. The complexity of the photoreceptor/RPE/choroid system and the strong interdependence of its components make it difficult to isolate a single causative event that triggers the disease. Nevertheless, there is agreement among researchers that the disease is caused by some RPE dysfunction, precipitated by an inherited susceptibility and/or environmental exposure. All of the above-mentioned entities (macular pigment, lipofuscin, melanin, Bruch’s membrane, and the choroidal blood supply) may have roles in causing, exacerbating, or preventing the degeneration of the retina in AMD. Non-invasive methods are required to visualize and measure all of them.

  • Age-related increases in lipofuscin in the RPE may compromise the metabolic activity of the RPE and increase photochemical damage.
  • Because melanin tends to decrease with age (similar to the graying of hair), there may be a reduction of the protection given by RPE melanin.
  • Low macular pigment levels in the retina may predispose a retina to the damage of AMD.
  • Accumulation of extracellular materials between the RPE and Bruch’s membrane leads to deposits (called drusen) large enough to be visible by ophthalmoscopy; they represent the hallmark of the early-stage AMD. These deposits may distort the RPE and cause a breakdown in the tight junctions between the RPE cells, causing blood vessels and fluids to invade the retina.
  • Alteration in choroidal blood flow may also be important.


Research Interests

  • Retinal pigmented epithelium (RPE) lipofuscin in aging
  • Age-related macular degeneration (AMD)
  • Stargardt disease

RPE Lipofuscin in Aging and Age-Related Macular Degeneration

Lipofuscin (LF) accumulates throughout life in the retinal pigmented epithelium (RPE) as a result of oxidative damage to photoreceptor membranes. During old age, lipofuscin - trapped in lysosomes - is a major cellular constituent and can occupy as much as 25% of the free cytoplasmic space. Growing evidence indicates that these large quantities of LF in the RPE may have a negative impact on retinal health, and possibly contribute to pathogenesis of age-related macular degeneration (AMD). An important constituent of LF is the red-emitting fluorophore referred to as A2-E (to indicate that it is a molecule derived from 2 molecules of vitamin A covalently bound to ethanolamine). A2E has been shown to inhibit lysosomal digestion of proteins, to induce lysis of the lysosomal membrane causing release of lysosomal markers into the cytoplasm, and to act as photosensitizer in blue light generating free radicals within the RPE cell.

LF was measured in vivo by spectrofluorometry; excitation and emission spectra of the autofluorescence of the retina were measured in 2°-diameter areas of the retina. The dominant fluorophore exhibited spectral characteristics that were consistent with A2-E. However, secondary fluorescence sources, identified as originating from Bruch's membrane, drusen in AMD, and the vitreous interfered. To measure A2-E, Dr. Delori and colleagues minimized the contributions of the secondary sources by using an excitation wavelength at 550 nm and by detecting the fluorescence above 640 nm. This had the added advantage that the measurement in the fovea was not affected by macular pigment absorption. They also developed a new device to image the retinal autofluorescence in order to study the distribution of LF at and around pathological tissue. The combination of the spectral and spatial information provided a powerful tool to identify and quantify LF and other fluorophores and to understand better the phenomena associated with changes in autofluorescence.

In the normal population, LF increases significantly with age, being approximately 4 times higher in the 7th than in the 2nd decade of age. LF decreases slightly in the 8th and 9th decade of age, probably as a result of photoreceptor loss and/or early subclinical degeneration. The distribution of LF across the posterior pole is not significantly altered by age (foveal LF is about 60 % of LF at 7° from the fovea). In patients with clinical AMD, LF is lower than in normal subjects of the same age and the decrease becomes more pronounced as AMD progresses. The decreased LF levels in AMD are observed throughout the posterior pole, even in fundus areas with no visible drusen. In the last two years, Dr. Delori increasingly focused his work on the local relationships between LF levels and pathology in AMD, principally in regard to drusen, geographic atrophy, and hyperpigmentation.

Dr. Delori investigated the distribution of autofluorescence at the site of drusen and found that the center of individual drusen exhibited reduced LF, whereas the surround exhibited high LF in most cases. Hard and soft drusen with sizes between 60 and 175 mm exhibited the pattern most clearly. In more advanced disease the pattern became less distinct and foci of high and low fluorescence, not clearly associated with the loci of drusen, were increasingly seen. For well-defined profiles, he found that the integrated fluorescence was not significantly different from the surrounding background, which is consistent with a peripheral displacement of the overlying RPE cytoplasm and/or LF granules without an actual loss of the RPE. However, several drusen profiles exhibited significant decrease in total fluorescence, which may indicate incipient atrophy of the RPE. Dark autofluorescence foci without a distinct annulus were seen in advanced AMD and may mark areas were drusen have regressed. This would be consistent with histopathological findings that the RPE is attenuated at the site of faded soft drusen.

Areas of geographic atrophy were extremely well delineated in autofluorescence imaging by their very low fluorescence. The fluorescence spectra within the atrophy had the characteristics of choroidal collagen, with no or little evidence of LF. The borders around areas of geographic atrophy exhibited highly variable LF levels, including foci with high LF. These observations were supported by histopathology: the junctional zone of geographic atrophy contained irregular and enlarged RPE cells, which were densely packed with LF and melanolipofuscin granules. We hypothesize that the focally elevated LF at the edge of geographic atrophy identifies the region at greatest risk for new atrophy. In a current longitudinal study, we will determine whether the annual radial increase in atrophy is correlated with the fluorescence intensity along the perimeter of the atrophy, and whether the direction of maximum enlargement is predicted by the location of the brightest fluorescence along the perimeter.

There is growing evidence that hyperpigmentation is an indicator of high risk for disease progression, or neovascularization, in AMD. Dr. Delori and colleagues demonstrated that foci of hyperpigmentation exhibit higher autofluorescence than do the surrounding background, and that the fluorescence characteristics of hyperpigmentation are consistent with those of LF measured at neighboring sites in the same eye. In histopathology, hyperpigmentation corresponded to regions where hypertrophic, pigment-laden RPE formed multiple layers, or dispersed in the subretinal space and outer nuclear area in clumps of pigmented cells. Thus, hyperpigmentation may represent local accumulation of LF or melanolipofuscin. These observations raise the hypothesis that excess LF not only is a marker of active pathological processes, but actively contributes to the development of the pathology. Dr. Delori is currently following several AMD patients with hyperpigmentation to determine the spectral characteristics and whether either those characteristics or the absolute fluorescence levels are correlated with prognosis.


Selected Publications

Dr. Delori has published more than 70 peer-reviewed articles. Below is a list of selected publications. View his publications on PubMed.

  1. Sparrow JR, Duncker T, Woods R, Delori FC. Quantitative Fundus Autofluorescence in Best Vitelliform Macular Dystrophy: RPE Lipofuscin is not Increased in Non-Lesion Areas of Retina. Adv Exp Med Biol. 2016;854:285-90.
  2. Duncker T, Tsang SH, Lee W, Zernant J, Allikmets R, Delori FC, Sparrow JR. Quantitative fundus autofluorescence distinguishes ABCA4-associated and non-ABCA4-associated bull's-eye maculopathy. Ophthalmology. 2015 Feb;122(2):345-55.
  3. Burke TR, Duncker T, Woods RL, Greenberg JP, Zernant J, Tsang SH, Smith RT, Allikmets R, Sparrow JR, Delori FC. Quantitative fundus autofluorescence in recessive Stargardt disease. Invest Ophthalmol Vis Sci. 2014 May 1;55(5):2841-52.
  4. Greenberg JP, DunckerT, Woods RL,Smith RT,Sparrow JS, Delori FC. Quantitative Fundus Autofluorescence in Healthy Eyes. Invest Ophthal Vis Sci. 2013; 54:5684-5693.
  5. SparrowJR, BlonskaA, FlynnE, DunckerT,GreenbergJP, Secondi R, UedaK, Delori FC. Quantitative Fundus Autofluorescence in Mice. Correlation with HPLC Quantitation of RPE Lipofuscin and Measurement of Retina Outer Nuclear Layer Thickness. Invest Ophthal Vis Sci. 2013; 54: 2812-2820.