As a physicist and materials scientist, the brain has been a challenging organ for me to explore. During the last phase of my PhD, my beloved aunt-in-law was diagnosed with brain cancer and unfortunately passed away before I graduated. When I was interviewed by Professor Bob Langer for the postdoctoral position at the Koch Institute for Integrative Cancer Research, I was told that professors Langer, Cima, and Graybiel of MIT had an NIH grant for work on a brain implant. The timing was just right: I expressed my profound interest in this project and became the lead researcher. In the course of my postdoctoral research, I capitalized on my microfabrication skills in 2D platforms and converted my device design and fabrication talent into 3D neural platforms.

In Turkish culture, we serve Turkish coffee in tiny cups made of fine porcelain, carried on a tray. The tray provides stability to carry these tiny, fine-featured coffee cups and plates to our guests.

The same idea was applied to my thinking about my brain implant device structure. The miniaturized neural drug delivery system—which we call MiNDS—has multiple tiny device components, such as two fluidic channels connected to wireless micropumps for delivering nanoliters of drugs on demand, and an electrode to record neural activity for potential feedback control. These components are all thinner than a hair fiber and can’t be handled with bare hands.

The scale of these tiny, delicate device components is why I initially microfabricated a polymer tray to carry them all on a planar silicon substrate in 2D format. While the mechanical stability is provided by the polymer tray, I used my microfabrication tricks and lifted-off the entire device platform from the planar, rigid substrate and encased it in a round, flexible stainless steel needle. The resulting system then turns into a 3D platform to reach deep brain structures without the need of an extraneous guide tube to implant in the brain. Overall, MiNDS has a diameter of 200 um, slightly thicker than a hair fiber. Furthermore, MiNDS is scalable, with its  length modifiable according to the desired application. For instance, in trials with small animals, we used a small MiNDS with a length of 1cm, whereas for nonhuman primates we used a 10cm one.  

I have been working on this project in Professor Robert Langer’s research group since September 2014, continuing the work at the Media Lab starting  January 2017. I was fortunate to collaborate with Professor Michael Cima (Koch Institute) and Professor Ann Graybiel (McGovern Institute for Brain Research), both at MIT, and to work with talented engineers and neuroscientists in these research groups.

The biocompatible, remote controlled MiNDS permits dynamic neural adjustment with pinpoint spatial resolution and cell-type specificity. With dual chemical delivery channels and an electrode, microfabricated MiNDS can chemically modulate local neuronal activity and related behavioral changes in animal subjects while simultaneously recording neural activity to enable feedback control. With this method, we can achieve a decrease in systemic toxicity and therapy time within seconds.

Any brain disorders that can be treated with drug therapy could be explored with MiNDS. This technology could pave the way toward an adaptive, multimodal treatment for neurologic diseases—and eventually revolutionize therapy for patients. At the current stage, we would like to explore the underlying mechanisms of, and eventually treat, Parkinson’s disease.

MiNDS has been implanted in several small-animal brains, and its functionality tested for up to eight weeks. The wireless micropumps for on-demand delivery of nanoliters of drugs were also implanted. The micropumps allow for refilling of the drug reservoir, even while implanted, via a septum that can be penetrated using a 31-gauge needle.     

The multimodal MiNDS has a outer diameter of 200 um, only slightly thicker than a hair fiber. The overall system, including micropumps, MiNDS, and associated infusion connections, takes about two hours of surgery for insertion.

So far, we have conducted pre-clinical trials in small (rodent) and large (nonhuman primate) animal models. Clinical trials are not in place yet.

Yes. For example, another potential use of MiNDS could be for targeted delivery of chemotherapeutics to tumors in the body. Such a technique would provide delivery of higher doses without the associated systemic toxicity. MiNDS could also be used to deliver growth factors and stem cells to regions of significant cellular necrosis. For neurological and cardiovascular diseases, combining growth factor therapy with electrical stimuli might help regenerate electroactive cells.  

I would assume that we need 5-10 years to have this technology ready for use in clinics by doctors.

Yes, Ed Boyden and his team do brain implants with light. The customizable feature of our MiNDS could open new routes to deliver not only light but also chemicals and electricity to other organs (not limited only to the brain) with pinpoint spatiotemporal resolution. For instance, optogenetics of peripheral nerves together with electrical and chemical interfacing could be achieved through a single implant via a MiNDS.

In my Conformable Decoders research group at the MIT Media Lab, my students and I are also working on integrating small, conformable sensors into MiNDS to monitor real-time phenotypic parameters such as temperature, pressure, stiffness during drug infusions, and also an integrated tissue retrieval channel for obtaining brain biopsies. Such multimodal capabilities would allow for more in-depth investigation into the pathology of neurological conditions in vivo as well as a novel drug exploration system for pharmaceutical research and development.