“Optical Coherence Tomography — Past, Present, and Future”
This past Friday saw the sixth annual Glaucoma 360 New Horizons meeting, held at the Palace Hotel in San Francisco. This year’s Drs. Henry and Frederick Sutro Memorial lecture was given by Professor Joel Schuman, now the chairman of ophthalmology at NYU Langone Medical Center. Professor Schuman needs little introduction and is a fitting choice for what is becoming a prestigious keynote lecture.
Entitled “Optical Coherence Tomography – Past, Present and Future”, the talk was a fairly candid and very interesting overview of OCT from one of its inventors. Born in the laboratory of Professor James Fujimoto at MIT (working in high speed lasers), the motivation was to do a non-invasive tissue “optical biopsy” based on work done at Bell labs in time-gated picture ranging. The idea in time-gated ranging is that you can look at reflections at a particular point in space, or range, using an interferometer. This was demonstrated by an ATT logo that was covered in tape but still resolved through time-gated ranging, which in this example was gating just the light scattering from the plane in space lying behind the tape. With this technology, you could look, for example, beyond skin and depth resolve information beneath. In 1989 Dr. Schuman was doing a glaucoma fellowship at Massachusetts Eye & Ear Infirmary when he heard of the project using optical ranging to measure the thickness of the cornea. The project was on shaky ground for two reasons. Firstly, it took too long to measure the thickness; and secondly you don’t need to do it in photorefractive laser surgery, the intended application, as you know how much tissue is being taken off with every excimer laser pulse. As Dr. Schuman astutely pointed out, projects fail all the time. But as we all know, the lessons learned can often be applied elsewhere. In this case, it was Dr. Schuman’s insight to suggest that the light source’s wavelength was suitable for reaching the posterior pole, and it might therefore be possible to get a signal out of the retina. A true light bulb moment that has had profound effects! The head of the laser lab at that time was Professor Carmen Puliafito, who perhaps needs no introductions. And here is where Dr. Schuman was refreshingly candid. Describing Dr. Puliafito as larger than life character, Dr. Schuman tells how his idea was deemed as “not the greatest in the world”. Undeterred, Dr. Schuman visited the lab at MIT with a bag of cat eyes and worked with Professor David Huang to get an OCT depth-resolved signal, and a-scan, from them. It was Dr. Huang who had the invention to consider transverse scanning in order to create a 2d image, a b-scan, from the a-scans. This again had never been done and they were then able to create tomogram images of the retina. This resulted in Dr. Huang being the first author of the now seminal Science publication and OCT was established.
OCT uses low coherence light, so a broad bandwidth that facilitates high resolution imaging. Its inherent simplicity is part of its strength and from the first bread-boarded system built by Eric Swanson and coded by a retinal surgeon, Dr. Michael Hee, a prototype system evolved and was used clinically. This technology was licensed by John Moore who was then president of Humphrey-Zeiss (now Carl Zeiss Meditec), and the program was led internally by Jay Wei, who later went on to form Optovue Inc. The third iteration of OCT devices made by Zeiss, the Stratus, really caught on, in part due to supporting clinical trials, but mostly due to faster scanning, a normative database and the all important billing code that enabled ophthalmologists to afford the technology.
OCT is the most rapidly adopted technology in ophthalmology, drives clinical decision making in a number of key areas, and continues to evolve at pace. Looking at the economics, the global market for ophthalmic OCT reached $723 million in 2013, climbing to $794 in 2014. The forecast is an annual growth rate of 7.9% through 2019, setting it to reach $1.2 billion by that time. Of this, polarization sensitive OCT will grow to $28.7 million.
As Dr. Schuman pointed out, this is a great story as the initial investment was the federal government’s National Institute of Health funding supporting the research at MIT and collaborating labs. This investment in basic science has resulted in a technology that benefits the health of the American people, the mission of the NIH, and also the economy. It is common knowledge that this is an amazing success story and the NIH should be roundly applauded for supporting such impactful technologies as successes are never guaranteed, but without such funding of science only failure can be guaranteed.
Today’s OCT is spectral domain OCT (SD-OCT) that is slightly different to the time domain OCT, and has a number of advantages in terms of throughput, sensitivity and resolution. The speed increases have importantly facilitated the acquisition of fully three-dimensional volumes, which has a number of advantages in terms of clinical utility, reproducible measurements being the main one. Dr. Schuman presented work done by his lab then at Pittsburgh in the relationship between structure, as measured using peripapillary RNFL, and function measured with visual fields. This important work showed that above 75um mean thickness the relationship was weak, but below that it was quite strong. This “tipping point” showed evidence that structural loss precedes measurable functional loss, as had been predicted by Professor Harry Quigley and Professor Alfred Sommer from histology and disc photography assessment. The amount of measurable structural loss was around 17% before measureable functional loss occurred. Very interesting, their work also showed that the structural measures bottom out, such that there comes a point when you can only follow disease progression using the visual fields. If only structure is measured, it can therefore imply that the patient is stable, but they may still be progressing in a way that can only be seen in the visual fields. Longitudinal data has proved these scenarios in a broad set of patients. Note that this floor effect is based on what is measureable by the machine, so it could be an error in the segmentation that reduces the dynamic range of the OCT device.
Dr. Schuman also presented work from the Advanced Imaging for Glaucoma (AIG) study, that was led by Dr. David Huang. Here a number of factors were looked at in the study of glaucoma progression and they found that the OCT abnormality that predicts visual field conversion from suspects or those having preperimetric glaucoma was in the ganglion cell complex (GCC) in the macula. Specifically, it is the focal loss volume (GCC-FLV) parameter was the best predictor of glaucoma conversion. Significantly, there was a two year lag between detectable progression in this measurement before a detectable abnormality in visual fields. I emphasize this as in the US at least, every two year’s I have a visual field test when I visit my optometrist as there is a reimbursement code for this. So although my visual fields are good, and I have taken about six tests, I have never had an OCT because there is no reimbursement code for that unless I were deemed to be of high risk. Based on the clinically evidence, this should likely change in the future as very simply OCT detects progression earlier. Dr. Schuman also reported that GCC-FLV was also found to be the best parameter to follow for perimetric glaucoma.
In terms of what’s coming, Dr. Schuman finished by presenting some new findings in the field of OCT-Angiography (OCT-A). Very importantly, in terms of quantification, Dr. Schuman positioned the technology as being just at the tip of the iceberg in terms of our understanding. Incidentally, we’ve blogged elsewhere on this, partly underlining just why this is the case. The work presented was very recent with data from Dr. Huang’s lab. It showed that retinal perfusion defects from OCT-A in both the macula and also peripapillary were seen to correspond well to the area of abnormality in the visual fields in perimetric glaucoma and controls. In the macula it was shown that it is the superficial vascular plexus that was significantly lower in terms of vascular flow density than the intermediate or deeper plexuses when comparing the glaucomatous eyes to controls. This might be expected because this is where the damage occurs in glaucoma, as Dr. Schuman pointed out. In peripapillary it was the radial peripapillary capillary plexus that differed most across glaucomatous eyes and controls.
It’s worth pointing out here that these tools are in clinics throughout the world, only an understanding of what measurements can be derived and the clinical utility is in its early stages. So the data is there and needs to be mined in the same way Dr. Schuman and colleagues have shown how OCT can be used with visual fields and when it is most useful in patient screening and management. I contrast this somewhat with the search for new biomarkers for glaucoma that is part of the catalyst for the cure initiative. Here, and as Professor Goldberg discussed in his talk later in the day, the emphasis is in molecular biomarkers, or at least new biomarkers from imaging at the cellular level, using, for example, adaptive optics. It is planned that these go to clinical trial this year to see if they are useful and complementary perhaps to those that exist and are understood. The lead time to clinical utility for such modalities and tests strikes me as a long one when there is so much to be mined from the OCT data that is in abundance. The implication is that the candidate biomarkers will be in the clinic soon, but that will require quantification, so robust and understood analysis. This will require software and then better software (see Dr. Schuman’s hopes for the next 5 years for OCT), which may be hindered by development costs (again, see Dr. Schuman’s next 5 years bullets). To be clear, this research is world class and important, in particular for the study of neuroprotective agents. But I digress to note that that it is not as immediate as implied, in particular when there is so much still to gain from the somewhat ubiquitous OCT data.
I feel my point here is somewhat reinforced by Dr. Schuman’s conclusions. Namely that:
- Advanced ocular imaging enables visualization and quantification of new ophthalmic tissues and structures.
- This facilitates the detection and management of glaucoma and more traditional ocular diseases.
- And that it is this information that will ultimately lead to new discoveries on glaucoma pathogenesis, leading to new treatment options to preserve vision.
Lastly, here are the bullet points presented by Dr. Schuman relating to what changes we will see in OCT over the next 5 years:
- “Higher speed will enable en face OCT imaging and OCT angiography”
- “OCT-A will become a standard adjunct to structural OCT – enabling 3d visualization of retinal and choriocapillaris vasculature”
- “Software advances will enable advanced processing and quantitative assessment of 3D OCT data.” i.e., to do better in terms of the parameters that are evaluated as biomarkers of the disease.
- “Subtle changes in pathology will be measurable, enabling more accurate monitoring of disease progression and response to therapy”
- “Competition between manufacturers will drive innovation and reduce prices”
- “However, development costs, market limitations and reimbursement changes may limit progress”
Industry Panel Discussion: Glaucoma Devices and Pharmaceuticals
Somewhat anecdotal, but worth documenting all the same are the responses of the industry panel to what they think will be the next big things over periods of 2, 5 and 10 years. In particular, the question was phrased as “what’s the next transformative thing that is going to happen and what should people be investing in now given the long lead times incurred in this industry?” The context of course is as pertains to glaucoma. As an individual working in the world of diagnostics, I found some of the answers very interesting. Particularly as I think better software (see keynote above) will play an increasingly important role in glaucoma and other diseases in the coming years. It is worth noting the individual’s role in each organization ahead of reading about their thoughts. The replies were more hurried as we arrived toward the end, but in order of response, they are:
- Sun Pharma – Mark Jasek, Vice President of Medical and Scientific Affairs, Ophthalmic Business:
- 5 years / 10 years: the use of imaging and new biomarkers making the move away from IOP. i.e., better diagnostics driving new therapeutics.
- Shire — Susan Benton, Head of Business Development, Ophthalmics:
- transformational: advances in diagnostics and the potential for neuroprotection.
- Santen Pharmaceuticals Co., Ltd. — Mike Garanzini, Head of Marketing, Europe (Glaucoma, Dry Eye, Retina):
- 2 years: surgical glaucoma (MIGs and beyond).
- 5 years: drug delivery.
- 10 years: last answer cut off by moderator to “keep it secret”.
- Novartis Alcon Pharmaceuticals — Melissa Liew, Medical Unit Head, Ophthalmology:
- 2 years: MIGs changing the treatments paradigm.
- 5 years / 10 years: Silicon Valley’’s interest in healthcare. I.e., the likes of Google Amazon and Apple getting in on the act could be transformational.
- New World Medical – Rafael Chan, Chief Commercialization Officer:
- 2/5/10 years: for late stage disease: improving the patient experience through better surgical procedures/devices, better management technologies (e.g. bleb, IOP management), better access globally to quality care.
- Carl Zeiss Meditec — Angelo Rago, Head of Ophthalmic Diagnostics:
- 2/5 years: agrees with all previous comments.
- 10 years: robotic surgery and tele-treatment.
- Bausch + Lomb — Tracy M. Valorie, BS, MBA, SVP, GM of Ophthalmology Rx:
- 2/5 years: new products having impact on trabecular meshwork.
- 10 years: the effort being spent in trying to slow the disease down.
- Allergan, Inc. — Aziz Mottiwala, VP of Marketing, US Eye Care:
- transformational: products to the patient, tailored made to the given patient (personalized medicine).
- Alcon Surgical — Shawn O’Neil, Head, Sales and Marketing Surgical Glaucoma, U.S. Surgical Division:
- transformational: if we had diagnostics, real-time or inter-operative to identify what the right treatment is and what will be most effective for that patient.
- Abbott Medical Optics, Inc. — Jane Rady, DVP Business Development:
- transformational: how the medical community will define for the patient how new products and procedures are optimized.
- Aerie Pharmaceuticals — Thomas Mitro, President and COO:
- Transformational: look at disease modification, put the patient back in charge.