Every two months, North­eastern bio­engi­neering grad­uate stu­dent David Walsh’s 91-​​year-​​old grand­mother goes to the doctor to receive a drug injec­tion into her eyes. She has wet age-​​related mac­ular degen­er­a­tion. There is no cure, only this inva­sive, recur­ring treatment.

She wor­ries a lot because she goes in, they inject her, and she leaves, and since the effect of the drugs is so gradual, she doesn’t know if it’s working or not,” said Walsh. “She wor­ries that maybe she got too much drug, not enough drug, or if the drug is doing any­thing at all for her condition.”

David Walsh and his grand­mother, Alice Neves, who has age-​​related mac­ular degen­er­a­tion and was an inspi­ra­tion for Walsh’s research. Photo cour­tesy of David Walsh.

To solve this problem, Walsh is devel­oping a device that will pro­vide valu­able feed­back to patients such as his grand­mother and their clinicians.

As a member of asso­ciate pro­fessor of chem­ical engi­neering Shashi Murthy’s lab, Walsh helps design microflu­idic devices that use a single drop of blood or other bodily fluid to diag­nose a range of dis­eases. In work recently reported in the journal Lab On a Chip, Walsh and his col­leagues have cre­ated a device that mon­i­tors the effi­cacy of treat­ments for two eye dis­eases: age-​​related mac­ular degen­er­a­tion and dia­betic retinopathy.

These are dis­eases where little blood ves­sels grow in the back of the eye and can pop your retina off or obscure your vision,” Walsh explained. This, he said, is due to increased levels of a mol­e­c­ular bio­marker called vas­cular endothe­lial growth factor, or VEGF.

The stan­dard treat­ments for these two eye dis­eases involve using drugs that bind to VEGF, thereby blocking its ability to interact with the cel­lular envi­ron­ment and causing blood vessel growth. These drugs are injected directly into the eye typ­i­cally every four to six weeks, and often over the course of a life­time. Walsh said the inva­sive nature of treat­ment and lack of per­son­al­ized dosage raise patient risk and dis­com­fort. What’s more, clin­i­cians don’t usu­ally test to con­firm the effi­cacy of treatment.

That’s because the cur­rent diag­nostic protocol—ELISA, or enzyme-​​linked immunosor­bent assay—is extremely expen­sive and time con­suming. Unless there’s no vis­ible improve­ment after many injec­tions, it doesn’t make sense to do an ELISA test, Walsh explained.

That’s where his device is dif­ferent. Each time a patient under­goes a round of treat­ment, which includes a drug injec­tion directly into the eye, a clin­i­cian can also take a very small sample of eye fluid—which con­tains a plethora of information—and test it in house in fewer than 20 minutes.

The device works like a light switch, where the on-​​button is only acti­vated when VEGF is present. And the more VEGF, the brighter the light.

It’s made of two thin disks just a few inches wide, held together by double-​​sided sticky tape that’s been expertly etched into a spe­cific design. That design cre­ates an array of micro­scop­i­cally thin hollow chan­nels into which Walsh inserts three reagents of varying den­sity. The first is the densest and con­tains a pho­toac­tive sub­strate that glows in the pres­ence of a par­tic­ular enzyme. The second layer simply acts as a den­sity gra­dient, while the third layer is the ocular fluid sample that Walsh wants to test. This layer also con­tains the enzyme as well as some tiny dense beads. Both of these are linked to an anti­body spe­cific to VEGF, which allows them to form a com­plex with the biomarker.

Spin­ning the whole device at very high speed allows these dense com­plexes to migrate to the layer con­taining the sub­strate where the enzyme causes it to ignite. But if there’s no VEGF in the sample, the enzyme stays in the least dense layer and no light is produced.

The team col­lab­o­rated with clin­i­cians at the Duke Eye Center in Durham, North Car­olina, to obtain fluid sam­ples with which to test the device. It passed with soaring colors, dis­tin­guishing not only between patients with age-​​related mac­ular degen­er­a­tion or dia­betic retinopathy and patients with other dis­eases, but also between patients who have active and inac­tive forms of either of these two eye diseases.

The devices cost less than $1 and take min­utes to assemble and run. With this plat­form, a clin­i­cian wouldn’t need a com­pelling reason to test whether treat­ment for these two eye dis­eases is effective—instead, she could look at a patient’s VEGF levels every time he came in for treat­ment. Blood vessel growth without VEGF would be a strong—and early—sign that some­thing else was going on. What’s more, lack of reduc­tion of VEGF over a few treat­ments would indi­cate a non-​​responder and allow the clin­i­cian to switch to another drug or treatment.

Infor­ma­tion like this could not only help put Walsh’s grandmother—and patients like her—at ease. It could also help diag­nose more serious con­di­tions earlier.

Sep­a­rately, Walsh also recently earned a NSF Grad­uate Research Oppor­tu­ni­ties World­wide grant to per­form related ocular diag­nostic research at the KTH Royal Insti­tute of Tech­nology in Stock­holm, Sweden. He will begin that work in September.