Personalised therapies

Systemic and ophthalmic biomarkers may facilitate a better understanding of diabetic retinopathy

Dermot McGrath

Posted: Tuesday, May 1, 2018

Automated segmentation of a macular OCT image, showing several retinal layers along with the outer boundary of the choroid

Gabor Mark Somfai MD, PhD

Systemic and ophthalmic biomarkers may facilitate a better understanding of diabetic retinopathy (DR), and pave the way towards more personalised therapies to prevent vision loss in people with diabetes, according to Gabor Mark Somfai MD, PhD.

“There has been a lot of interesting research over the past few years into the mechanisms of diabetic microvascular complications. It is known by now that adipose cells are producing various pro-inflammatory cytokines that circulate in the blood and are responsible for insulin resistance. Vascular damage by the advanced glycation ultimately leads to diabetic macular oedema (DME) and proliferative diabetic retinopathy (PDR).

“To prevent this, I believe one day we might be able to determine the adequate and optimal preventive steps or therapy for these patients by looking at the pattern of their metabolism,” Dr Somfai told delegates attending the 8th EURETINA Winter Meeting in Budapest.

The duration of diabetes and the severity of hyperglycaemia are the two major risk factors for developing retinopathy, said Dr Somfai. According to the pivotal studies in diabetes, approximately 25% of patients with type 1 diabetes will develop retinopathy after five years, rising to 60% after 10 years and 80% after 15 years. For type 2 diabetes, 40% of patients with insulin control and 24% without insulin will have some form of retinopathy after five years, increasing to 84% with insulin and 53% without after 19 years.

“We need to bear in mind that control is the only controllable factor. Once diabetes is present, then the duration is less important than glycaemic control. A HbA1c blood glucose level of below 7% is recommended. In the United Kingdom Prospective Diabetes Study (UKPDS) study, each 1% reduction in HbA1c with intensive glucose therapies was associated with a 37% reduction in the risk of retinopathy, which is very significant,” he said.

Dr Somfai said that research by Dr Katja Hatz (Basel, Switzerland) presented at the ARVO in 2017 has shown that the total HbA1c load over the entire course of diabetes correlates highly with the DR status of the patient even after adjusting for age, sex and duration of diabetes.

“The total load shows a strong correlation with DR status and may one day be used as a marker of glycaemic control in patients,” he said. The concept of “metabolic memory” should also be borne in mind when treating diabetic patients, said Dr Somfai.

“Essentially, this means that the cells will remember the past glycaemic control for at least five years, whether it is good or bad. Bad glycaemic control will have an influence on cell metabolism for this period, so the patient needs to be advised that prior glucose control has sustained effects that persist even after return to more usual glycaemic control,” he said.

Blood pressure and cholesterol control should also be closely monitored to help reduce the risk of visual loss over the long term, added Dr Somfai.

Liquid biopsy of the vitreous could pave the way for a more personalised choice of intravitreal treatment for macular oedema, while systemic biomarkers could indirectly indicate the actual stage of diabetic retinopathy. Ophthalmic imaging biomarkers could also play an important role not only in the prediction but perhaps one day also in the choice of macular oedema treatment.

The structural and topographic information obtainable by standard optical coherence tomography (OCT) and the quantitative information of the macular microvasculature delivered by OCT angiography may also play an important role not only in patient counselling in terms of outcome expectations but also in therapeutic decision making.

In the not so distant future, Dr Somfai concluded, all these biomarkers could potentially be used in a novel approach together with big data analysis and machine learning, which would serve as rocket fuel for better clinical decisions.

Gabor Mark Somfai: