Interview with Bettina Willie,
Professor at McGill University and Shriners Hospital for Children in Montreal, Canada

Who are you and what is your relationship to OI?

I am Bettina Willie, and grew up in Utah, US. I earned a doctoral degree in Bioengineering from the University of Utah and I did my first postdoctoral training at the Institute of Orthopaedic Research and Biomechanics at the University of Ulm in Germany. I then did a second postdoc at the Hospital for Special Surgery, in New York City. I maintained my connection to Germany by working as a group leader in the Julius Wolff Institute at the Charité – Universitätsmedizin Berlin for nearly eight years. My research was then focused on examining how mechanical loading influenced bone formation and healing during skeletal aging and in rare bone disorders. Since 2015, I have worked as a Professor at McGill University and Shriners Hospital for Children in Montreal, Canada. There I decided to start a research program on OI myself. I use multidisciplinary approaches, including high-resolution imaging to find out how to treat bones that break easily.

Some of my research involves scanning people with OI using high-resolution imaging, called high resolution peripheral quantitative computed tomography (HR-pQCT). I served as the central reader for a recently completed Phase 2b, multinational, double-blind, dose-finding study in adults with Type I, III or IV OI treated with a new drug called Setrusumab. As the central reader, I was in charge of determining how we scan the participants, analyzing all the HR-pQCT data, and helping the 13 clinical scanning sites in the UK, France, Denmark, Canada and the US when they had questions. As part of this study, we also determined how reproducible the HR-pQCT scanning was in people with OI, which was published in the medical journal Bone (Mikolajewicz et al. 2021). My laboratory also developed different methods of scanning people with OI over time to find the same volume of bone and we tested how reproducible these methods were (Tabatabaei et al JBMR 2022). I make all of my lab’s methods and software codes publicly available at a website called Github, so that other scientists can freely use them.

Together with Dr. Frank Rauch and Dr. Francis Glorieux at the Shriners, I am currently performing the first study that uses HR-pQCT to scan children with OI over time to see how their bone structure changes. I have learned a lot from this study in terms of various challenges scanning and analyzing data from kids with OI as well as children in general. Since the bones of children are growing, it is challenging to image the same volume of bone at the beginning of the experiment and at one year follow up. My lab will determine if image analysis methods (called imaged registration methods) that are commonly used in adults can also be implemented in children. In addition, although there are standard imaging protocols established for adults, there is no consensus on where one should scan the bone of children. This has led to difficulty in comparing HR-pQCT data from different studies. There is a pressing need to standardize HR-pQCT scanning and data analysis methods in children.

I am also interested in learning why bones from people with OI break so easily. Our skeletons adapt to increased mechanical loading by forming more bone. For example, tennis players have denser bone in their playing arm than their nonplaying arm. A lack of mechanical loading can also lead to reduced bone mass. When a person does not get enough physical activity, the skeleton adapts by removing bone tissue. This often occurs with astronauts, individuals with paralysis or people who are bedridden for long periods of time. What’s more, bones have a reduced ability to adapt to mechanical loading as people age. Research shows exercises that promote bone formation in young people aren’t as effective in middle-aged or elderly people. But the molecular mechanisms involved in this altered response with aging are unclear. There is some evidence that the bones of people with OI may also have an altered response to mechanical loading. Dr. Frank Rauch and I recently received a five-year grant from the Canadian Institutes of Health Research to examine the bone formation response to mechanical loading in mice with OI.

What do you do when you are not working?

I am married with two sons. Both of my children have health issues, which has motivated me even more to focus my research on a disease that affects children in particular. I relax by spending time with my family, working in my garden, reading mystery novels, or watching the Toronto Blue Jays baseball team.

What it HR-pQCT?

HR-pQCT is a powerful imaging method that gives information on the bone’s structure and mineral density in the arms and legs. This allows us to estimate the bone’s strength and ability to resist fracture.

What is the difference between imaging methods when it comes to measuring bone density/quality in OI?

An X-ray is like a shadow of the bone that provides some qualitative information, including if the bone is fractured. DXA allows some quantitative information about the bone’s density. A normal CT scan allows for a 3-dimensional image of the bone, but bone density cannot be measured. To measure bone density, quantitative CT (QCT) must be performed. HR-pQCT is a form of QCT, which is limited to the peripheral bones (ie. arms and legs), but has a higher image resolution compared to QCT, enabling the assessment of the bone’s structure.

What are the best imaging methods to diagnose fractures in OI?

Fractures are usually easy to diagnose on X-rays. In some cases, it is necessary to use CT, MRI or DXA.

What kind of imaging has the lower/higher radiation doses?

HR-pQCT is a low radiation dose method; the effective radiation dose from a standard scan at the distal radius or tibia is 3–5 μSv. This is lower compared to other common medical imaging techniques. For example, a hip scan using dual-energy X-ray absorptiometry (DXA) has an effective dose of approximately 9 μSv, a standard chest X-ray approximately 100 μSv, and a hip CT scan 2000–3000 μSv. The amount of natural radiation that children would have during a day at home in North America is an effective dose of 8 μSv.

Why are there so few HR-pQCT machines worldwide?

HR-pQCT were introduced relatively recently; the first-generation scanners were introduced in 2004 and the second-generation scanners in 2014. Also, they are a relatively costly device priced at approximately $500,000 US.

What are the opportunities and challenges of HR-pQCT in OI?

Some challenges are that not many studies have been done in healthy children, so we do not know exactly how different the results are in children with OI. Also, it can be more difficult to position the individual’s arm or leg within the device, especially for people with more severe OI types, who have shorter or deformed limbs. The opportunity of HR-pQCT is to calculate bone strength better than with other techniques.

How can this be solved with standardization?

The expert teams that work on a similar HR-pQCT topic can collaborate to share their methodologies and data. This enables an evidence-based comparison of different methods performing a similar task, which could facilitate us reaching an agreement. These collaborations can result in consensus publications that provide recommendation for the community, which will also increase the impact of the work that each group has done.

Are HR-pQCT parameters predictive of fractures?

Currently, DXA is the most common imaging method used to assess bone mineral density and predict fracture risk. However, it is far from perfect, and many people still have bone fractures despite scoring high on their DXA bone scan. HR-pQCT allows us to measure the bone density more accurately than with DXA and it allows us to measure the bone’s structure and strength. Together with other experts in the field, I performed a review of all the studies that had been conducted using HR-pQCT (called a meta-analysis).

We found that HR-pQCT measures of bone density, structure, and strength were significantly lower in people with a prior fracture (Mikolajewicz et al. JBMR 2020). Another group calling themselves the Bone Microarchitecture International Consortium (BoMIC), also looked at a large number of previous studies using HR-pQCT and they found that bone density, structure and strength measured using HR-pQCT were important predictors of fracture and they were able to predict fractures better than only using bone mineral density measured by DXA (Samelson et al. Lancet 2018).

In what kind of research projects have HR-pQCT been used as an outcome measure?

The HR-pQCT scanner was first designed for osteoporosis research in adults. However, the use of HR-pQCT has not been limited to osteoporosis. An increasing number of studies are being performed to examine a certain disorder or a particular drug treatment over time.

You encourage professionals who work with OI & HR-pQCT to collaborate more?

I highly encourage the experts working with HR-pQCT to collaborate to facilitate the standardization of HR-pQCT scanning techniques in children and how they report methods. I am open to collaborating with other experts on these topics as well as share any of our methods.

Editor’s note: Do you work with HRpQCT in children and is interested in standardizing HR-pQCT scanning methods in kids? Bettina would like to organize an online meeting with stakeholders to come to a consensus on HR-pQCT scanning parameters for children, so that data can be better compared between studies. The goal would then be to write a consensus article outlining these scanning parameters. Her email address is bwillie@shriners.mcgill.ca

This article was first published in OIFE Magazine 3-2022.