Robots helping cancer researchers produce better, faster results

Scientists are using robots in multiple disciplines of cancer research such as screening, radiation therapy, biopsy, surgery, and many other crucial fields to help beat a disease that has touched and affected so many lives.
By Tanya M. Anandan February 3, 2017

Personalized medicine is one of the biggest trends in cancer research and treatment. Robotics is one of the enabling technologies that are helping scientists are working around the clock in multiple disciplines to find ways to beat a disease that has touched and affected so many lives.

For example, Notable Labs is building on the idea of personalized medicine and using a patient’s own cancer cells to guide therapy. The challenge has been to create a reliable screening process for the thousands of possible drug combinations that might be effective for each patient’s unique tumor.

Previous attempts that have failed have been with very manual systems, like what you would think of with scientists in labs with pipettes," said Matt De Silva, cofounder of Notable Labs, a biotech startup in San Francisco, Calif. "The biggest issue with that is the reliability and reproducibility of the data. Robotics and automation provide a reliable way to increase the reproducibility of the experiments."

Notable Labs screens 5,000 to 10,000 drugs and drug combinations against a cancer patient’s cells, and has the ability to screen up to 30,000 combinations. A feat impossible without robotics and automation.

De Silva cofounded Notable Labs in 2014 after his father was diagnosed with glioblastoma, a highly aggressive brain tumor. Frustrated by the lack of treatment options available to patients with brain cancers, De Silva embarked on a journey, one in which many patient advocates find themselves as they maneuver the maze of often disparate data on cancer research.

"The reason I felt compelled to start the company was because the treatment options that I was looking at for my dad were already FDA-approved. The only thing that was standing in the way of getting those options to him was the scientific rationale for why they might be useful against his specific cancer. Robotics and automation, and some really cutting-edge biology and science, and mixing all of those together, has made that a possibility."

Even after losing his father, De Silva continues his mission to change the way cancer is treated by repositioning treatment as a patient-centered service. Notable Labs has refined its process and with a better understanding of its clinical implications, shifted its focus to leukemia, or blood cancers, particularly acute myeloid leukemia (AML).

AML starts in the bone marrow and typically moves quickly into the blood. De Silva said they found it’s easier to obtain large numbers of cells from leukemia patients relative to the very few number of cells available from patients with brain tumors.

Figure 1: Robotic lab automation system helps accelerate personalized cancer treatment by testing thousands of drug combinations against a patient's cancer cells. Courtesy: Notable Labs, RIA"Each test that we run is a unique mixture of different cocktails of drugs customized to the patient’s needs clinically," De Silva said.

The entire process, from the time the lab receives the patient sample to when a report of the results is sent to the patient’s oncologist, typically takes 3 days.

"The ability to test the drugs on the cells the same day that you get them, and be able to do all of these experiments quickly, is really important from a biological perspective," De Silva said. "Many previous attempts involved taking cells from a patient and growing them in a lab, and then subjecting them to drugs. The problem with the ‘growing’ step is that you basically change the biology of the cancer cells. They began to adapt to being grown in the lab and behave less like they would in a patient.

De Silva said they are running at a much higher throughput, which means they can conduct more experiments per patient sample. This allows Notable Labs to test more drugs than just the obvious chemotherapies. FDA-approved drugs shown to have anti-cancer evidence include those for treating high cholesterol, diabetes, and heart disease. Even anti-depressants are on the drug panel.

Robots producing faster results

"Before we had the workcell, it took the entire day for one engineer and one scientist to handle 120 plates," said Transon Nguyen, Lead Engineer at Notable Labs. "Now, with dynamic scheduling, our robot is able to keep track of 20 different plates at all different stages in their lifetime and able to move between all the different instruments much more reliably than anyone could do manually."

He said that a day-long manual process is now reduced to about 30 seconds, the time it takes for one scientist to prep the workcell. The robot, interacting with the lab instruments and scheduling software, does the rest.

Figure 2: Collaborative robot is designed to work safely side by side with its coworkers in the lab. Courtesy: Precise Automation, RIAAt any given time there may be 3 to 5 processes happening simultaneously. Imagine trying to keep track of all that activity. This brings the concept of lights-out manufacturing to the lab. Nguyen said they regularly run automated processes overnight and over the weekend.

"Particularly with large screens, we’ll kick them off in the evening and they’ll run throughout the night. We’ll come in the morning and our screens will be done."

Collaborative robots in the lab

Biosero, a lab automation system provider and integrator, helped Notable Labs bring together all the components of the workflow, including robotics, instrumentation, and software integration.

The robot at the center of the all the action is the PreciseFlex Sample Handler (pictured), which is a lightweight collaborative robot made by Precise Automation Inc. in Fremont, California. This 4-axis SCARA robot with integrated servo gripper is designed for benchtop applications where ease-of-use, safety, and space requirements are critical.

The robot will not injure a user or damage instrumentation even if it collides with them at full speed. This eliminates the need for expensive safety shields and allows the robot to safely operate side by side with lab personnel.

The robot is also easy to teach. The user can move it by hand to the start and end positions and let the robot controller handle the rest. There’s no teach pendant, which is common with many industrial robots.

"Robotics technology has progressed to the point where they can embed the robot into the software and make it transparent to the user," said Mike Ouren, sales engineer for Precise Automation. "You can have a small room of equipment, a robot, and some really clever scientists to make all that stuff work, and you can begin to experiment with your idea at a much lower cost than it would have been 5 to 10 years ago," he continues. "That has enabled smaller organizations to do this kind of cancer treatment research."

Nguyen said the stakes are high. "You can’t make a mistake because that could mean losing a patient sample. But we still have to maintain speed and flexibility. We push most of the instruments on our platform to their breaking point many times."

Moving liquids with sound

Figure 3: Six-axis robot handles thousands of samples a day, transferring microtiter plates between lab stations with speed and accuracy. Courtesy: Stäubli, RIAAnother critical component of the process is an acoustic liquid handling device. Acoustic liquid handling technology uses a focused ultrasonic wave that ejects a nanoliter-sized droplet upwards into the well plate containing the patient’s cancer cells.

"It allows you to transfer very small concentrated amounts of drug, which makes the economics of this very attractive," Nguyen said. "But it also lets us mix different combinations much easier than with a traditional pipetting system."

Acoustic droplet ejection (ADE), or the transfer of liquids via sound waves, eliminates the need for pipette tips and the risk of cross-contamination. It also reduces the amount of patient sample required.

"When I say that we do much higher throughput in terms of the number of experiments that we run, it’s in part because we’ve been able to drastically miniaturize the amount of cells and the size of the experiment," De Silva said. "If you think of the number of cells that you get from a patient as being fixed, then being able to miniaturize the experiment means you have more shots on goal to find a treatment option. It’s hugely important for the success of this approach."

From drug screening to drug development

Clinical trials are underway at Princess Margaret Cancer Center in Toronto and the University of Florida in Gainesville. De Silva explains that these are not typical clinical trials. They are evaluating the feasibility of trying to personalize treatment for patients who otherwise don’t have treatment options.

"We’re really testing whether we can do this with the clinicians in real time. It’s a way to evaluate and improve the process we’ve developed."

He said that someday it will be as simple as a test that your doctor orders to check your cholesterol levels. De Silva said the approach also has implications for drug development.

"The promise of personalized medicine is to start to use these tools to predict which of the patients are going to be the strongest responders to a given drug or drug combination. Then use that information to start developing drugs that have a much stronger effect for the right patients," De Silva said. "If you take the data that you’re generating from each of these individuals with a given kind of cancer, and you look for patterns across those patients, then you use both the data and the lab itself to help pharma and biotech companies evaluate where their drugs are going to work, or which patient populations they should target. They are doing this already, but they’re doing it with genomics. Our lab allows them to physically test the drug on the cells."

In the future, De Silva hopes to use their approach for research and development with drugs that are not yet FDA-approved. In the meantime, once the original clinical trials are complete, Notable Labs’ main push is to make their service available to more cancer patients.

Making the impossible possible

High-throughput drug screening for cancer treatment taps into the innate advantages of robots. Their error-free repeatability, speed, and capacity to work 24/7 allows them to accomplish in days, even hours, what a room full of lab technicians and scientists couldn’t hope to achieve over a span of years.

"The robot is an extension of the scientist doing the work that they can’t do, and even if they could, wouldn’t want to do," Ouren said.

Harnessing our DNA

While robots are not new to big pharma companies and large research organizations, the automation bug is just starting to catch on with smaller labs.

"The big growth in robotics will come from small to medium-sized labs," Ouren said. "What’s now available with a robot and software combined, what they can get as an integrated solution and the versatility they can get, is way beyond what it was years ago."

That’s bad news for cancer. Robot-armed labs of all sizes, with different ideas and approaches, attacking cancer from all directions with targeted therapies. That’s more shots on goal. The more, the better.

According to Cleveland Clinic, liquid biopsies for early detection of cancer is one of the top 10 medical innovations expected to transform healthcare in 2017. Labs will need to ramp up their robot armies to handle all the blood samples.

DNA sequencing is getting a boost from flexible automation. Fully automated genetic testing labs have robots on the front lines working 24 hours a day testing our DNA for cancer risk factors. 

Robotic radiation therapy

Using robots to deliver radiation treatment is already commonplace in many large hospitals and cancer treatment centers. The Accuray CyberKnife System uses a KUKA robot to manipulate a linear accelerator (LINAC) around the patient to precisely deliver high doses of radiation to tumors.

In ongoing studies, researchers are using nimble collaborative robots for intraoperative radiation therapy. 

Corey Ryan, manager of medical robotics for KUKA Robotics Corporation in Shelby Township, Mich., discussed some of the technological advances that are helping surgeons in their work. "Normally, a surgeon has to estimate the right position and trajectory of the treatment guide based on an X-ray image," he continues. "However, the robot allows the surgeon to look at the X-ray and exactly plan the position and trajectory of the guide (relative to the patients known anatomy) by commanding the robot to go to that exact spot at a predetermined angle. The surgeon then does the insertion and drilling manually, using the guide. Finally, radiation is delivered by the robot into the exact treatment spot."

Robots are also used for patient positioning in proton therapy centers, where patients have to be precisely positioned and moved under the fixed proton beam.

Robotic biopsy and surgery

Robots bring superhuman qualities to already skilled surgeons, improving precision, minimizing damage to healthy tissues, and resulting in faster recovery times. Robots may soon be helping physicians optimally position interventional needles for biopsies, an often time-consuming procedure fraught with risks when done manually. Robots are also used for minimally invasive thoracoscopic, cardiac, urological, and gynecologic surgery. 

They say the ‘cure’ is sometimes worse than the disease. Let’s hope that personalized cancer therapy can make this a thing of the past and eventually lead to the right cures for all cancers.

Tanya M. Anandan is contributing editor for the Robotic Industries Association (RIA) and Robotics Online. RIA is a not-for-profit trade association dedicated to improving the regional, national and global competitiveness of the North American manufacturing and service sectors through robotics and related automation. This article originally appeared on the RIA website. The RIA is a part of the Association for Advancing Automation (A3). A3 is a CFE Media content partner. Edited by Chris Vavra, production editor, Control Engineering, CFE Media, cvavra@cfemedia.com.

ONLINE extra

See additional stories about other robotics developments linked below.

For more information about upcoming A3 events including A3 Business Forum and Automate, visit: www.robotics.org