Steven Cummings, M.D., featuring Cheryl Ackert-Bicknell, Ph.D.

In this ASBMR 40th Anniversary Podcast episode, osteoporosis pioneer Dr. Stephen Cummings shares the stories behind some of the field’s most important advances. From the landmark Study of Osteoporotic Fractures (SOF) and the rise of bone densitometry to the origins of FRAX®, T-scores, osteopenia, and vertebral fracture assessment, Dr. Cummings reflects on the research that reshaped fracture prediction and osteoporosis care. He also offers valuable advice for the next generation on the importance of rigorous study design, innovation, and understanding the history behind today’s clinical practice.

Transcript

SPEAKER_00 0:08

And Professor Stephen Cummings. It's a great pleasure to welcome you to the ASBMR 40th Anniversary Podcast Series. I have personally been reading your studies and referencing your studies for an extremely long time. So I've been looking forward to this interview for weeks and weeks and weeks, and it's been great preparing for this interview with you. So welcome and thank you.

SPEAKER_01 0:31

Well, thank you very much for the opportunity to tell the stories.

SPEAKER_00 0:36

As we were talking and planning about the many, many things that you have done in your career and the legacy and impact that you've had on our field. The one thing that we talked about was the early days of densitometry. And I've been trying to teach this to my grad students and residents, but I was hoping that you would take this opportunity to teach the whole next generation because it's such an important topic.

SPEAKER_01 1:02

Sure, great. I'm happy to. The story of densitometry goes way back into the 1980s before we had DEXA or even single energy absorptiometry. And uh when uh so when they were first developed, people were doing case and control studies comparing people with fractures or not, which tend to give biased answers. And so there was uh I I criticized that kind of research for a while. And uh, in order to resolve the predictive value of uh densitometry, you needed a study that was large enough, rigor enough, prospective, in order to test the association between densitometry and what you care about, which is the risk of fractures. And so we started the study of osteoporotic fractures for that purpose. 9,600 women who we enrolled at four centers measured bone density peripheral sites and then followed them up for fractures of various types. And over the course of a number of years, we showed that densitometry at those sites predicted diverse fractures and that all those sites were the same. But then an innovation happened where we were able to use dual X-ray absorption to get in to look at the spine and the hip bones more accurately. And so we adopted that innovation in the SAW study, and using the same large population in perspective design, we we demonstrated that in fact bone density at the hip was really superior, was a better predictor of hip fracture, and even a somewhat predictor, greater predictor of uh vertebral fractures. That led to the adoption of femoral neck bone density as the standard for the field, including enrollment in clinical trials, and I think in use its use in frags, its use in you know in the clinic, and I think that that was one of the major contributions of our SOS study. Uh then uh in addition to that, I'll add that as we were doing that, I also had one foot in the world of geriatrics and general internal medicine, and I had interviewed a number of a lot of patients with hip fractures, come to the understanding that hip fractures were not just a bone density problem. Hip fractures in particular had a lot of other influences like a high risk of falls and weakness and impaired vision and things that that happened with aging. And we showed in SOF by looking at those geriatric risk factors that they were just as important as bone density. And that means, I think that led to a recognition, you know, that hip fractures are not just a bone density problem, but something that, you know, uh would benefit by attention to other factors that lead to the fracture. And those are things like falls and prevention of falls, impaired vision. And that we we showed that though the bone density and those risk factors had independent and very similar effects on the risk of fractures. And that meant that the best way of assessing risk would be to combine risk factors and bone density in predictive models. I believe that really is part of the foundation of what we now use as fracks. The genius of fracks has been to assess both risk factors very simply from large databases and bone density, guess what, at the hip, and putting those two together to show that uh that risk factors and bone density are important to the genesis of fractures.

SPEAKER_00 4:52

There's more to tell about that story, but well, I'd love to hear more about the genesis of FROCs. Uh, right now, FROCS doesn't, the FROCS 1 doesn't really contain any information about fall risks, and the new generations of FROCs have started to expand on that topic a little bit more. So please, I'd love to hear more of your insight about the expansion and change of FROCS because I use it when I teach all the time.

SPEAKER_01 5:18

I I think it's a great tool, and I credit John Keynes with the vision of making this a widely available tool. And what they did, FRACs drew what was available from a number of cohort studies to build the first model that was successful. But it's not really in a position to do some kinds of uh risk factors that we did in our studies, for example, strength and walking speed and assessments which you could do about disability and frailty, all of which are important to fractures. But you know, that would make it a little more clumsy to use. And the fact that it's streamlined with a few risk factors that are easily recorded by patients, I think is its uh you know, its utility. Uh just trying to understand the fractures and what contributes to that. That's really not been Prax's purpose. That's something that other big studies have done in order to understand how you can prevent fractures and hip fractures in ways other than just changing bone density.

SPEAKER_00 6:24

I'd like to circle back to conversations that we had, the beginning of planning all of this and the definition of the T-score and where that's a fascinating study that not many people know about because it wasn't really described, it wasn't written about.

SPEAKER_01 6:43

I wrote about it in a in a uh in an article called Osteopenia, where I interviewed all many of the people who were at a critical meeting. The meeting was held in Geneva and it involved maybe, I don't know, maybe 10 people or so who were self-selected to get together to answer the question now. And we're getting it on our patients. What's normal? What's normal, what's abnormal? What should we call osteoporosis? What should we call, you know, uh what should we do with all this? And it wasn't clear mainly because most importantly, there is a continuous smooth association between bone density and fracture risk. And in that case, where do you cut it? Where on that smooth curve do you draw the line? And uh whenever that happens, including in the worlds of hypertension and diabetes, a committee gets involved. A committee makes a decision about where to draw the line. And in this case, the story is this self-appointed group, the committee, stared at that curve, and on the last day, I'm told, you know, they were kind of tired of arguing about it, and they'd been enjoying a bit of wine. And so one member of that group looking at this curve uh decided to draw the line at a T-score of minus 2.5. Why a T-score? That's another discussion, but that minus 2.5, because there were more women, you know, who were included in osteoporosis than if you drew it at 3. And you had too many women if you drew it at minus 2. And so I think it was Joe Melton that came up with the idea: oh, great, if we draw it at minus 2.5, 15% of women would be counted as osteoporotic. Well, that's great. That's the lifetime risk of hip fracture. Well, those two things are not really related to each other. But it was okay, you had to have a number, and that was the number that the committee agreed on, and it rapidly, because of the need for it, took on a life of its own. And somehow, and it's not clear how, somebody in that group then came up with the idea that we needed the population to be more concerned about bones. So let's add a cut off of a minus 1.0. And that was inserted in the report. Somehow, I don't know how. But in doing that and calling minus 1.0 osteopenia, about 50% of women over the age of 50 got a term, got a disease, and got something that still worries women today. And the first thing I tell women with osteopenia who are worried, hey, congratulations, you're normal. And this minus one was really a committee decision. It's not a disease. It's you know, it's just a word that a committee came up with. Don't worry. We care about your risk, not about osteopenia.

SPEAKER_00 9:47

I think it's very important that the next generation hears these stories and understands because right now these are definitions that we got taught in books. So thank you so much for sharing all of this.

SPEAKER_01 10:00

That goes a little bit further than that. That's important, that that is because there was no other number, it began to be used for enrollment into clinical trials. Women got into trials because they had that bone density level. And when that was used in trials, that became the indication for treating. So if you had osteoporosis by that T-score, God, where did that come from? I don't know. But since you had that level, you qualified, you had a diagnosis, and you could be reimbursed for prescription of the drug. So it had it really has had profound implications. It's been applied to other measurements. I don't know why. There's no basis for that application except that there was a consensus. So this, I think it's really important for people to understand, you know, the the arbitrariness, particularly of osteopenia, and the value instead, I think, of estimating fracture risk.

SPEAKER_00 10:57

I agree. Thank you so much. Another topic that we discussed was the origin of vertebral fracture assessment. And SOF really played an important role in this topic. So I was hoping you could tell some of the stories about what is vertebral fracture and what the role that your work and SOF played in all of that.

SPEAKER_01 11:21

Great. No, thank you. At the time SOF was funded, or when we got started, vertebral fractures were defined by a radiologist or maybe endocrinologist looking at a film and it looks like one of the vertebral bodies was squashed and ugly. And you could call that a fracture because it obviously was. But when you start doing really large studies like the SOF study, 9,600 women needing, you know, needing a definition of vertebral fractures, you can't do it with a radio a radiologist. So we needed another way that could be done in a more practical fashion. And at that time, a group in Hawaii that you have to credit, Dr. Woznich and Dr. Phil Ross, had been using a backlit board to put the films on and then put points, six points around each vertebral body. And they and then others began to define vertebral fractures based on how much, you know, how much, you know, how much height loss there had been based on those points. And it, you know, it you know, it served our purposes well for for a while, but then it became controversial. People said, well, some of these abnormalities on did on films may just be due to an endpoint compression, not really a fracture. And we found that the you know, this the decrease in anterior height was really quite common and probably normal. And so that you know, the fact that, you know, so those flaws, those limitations of the morphometric definition led to uh something that Dr. Harry Janant did, what I think was also important. He began to rate the vertebral fractures not only on height, but by their appearance, indicating uh by the famous diagrams of vertebral bodies being squashed to a different degree, that there were grades of vertebral fracture, and here was a way of melding what the what an expert radiologist sees with the ability to process large numbers of films for uh entrance into clinical trials as endpoints into large clinical trials. It made it it made the world of clinical trials feasible to do. And then Murray the Soft study with that placement of six points feasible to study the epidemiology of vertebral fractures. And the controversy still goes on.

SPEAKER_00 13:52

Yeah, it it certainly does. It's it's still a very active point of controversy.

SPEAKER_01 13:58

Richard Estell defined something called ABQ in diff as different from what Richard, what Dr. Jeanette described as SQ. And they both argued about which one is the gold standard. And I think now we've realized that in clinical trials that doesn't matter so much. What really matters, and probably in clinic as well, is whether that's changed. If you see a decrease, you know, that if that decrease happens, something happened to the vertebral body. It's it is that's that's a fracture. And that's so uh that that uh realization I think has supported clinical trials demonstration that they prevent vertebral fractures.

SPEAKER_00 14:45

Can you expand a little bit more about like this was instrumental for a laundronate?

SPEAKER_01 14:51

Oh yes. Uh we actually we our group proposed to Merck that we undertake a study uh with fractures as an endpoint for laindronate. And um we we uh we proposed that because SOF had an infrastructure that could recruit thousands of patients, and we had demonstrated that we could uh measure vertebral dimensions and therefore vertebral fractures in thousands of women, and that that ability using using morphometry to to do those kinds of large studies necessary for you know for the defining the fracture benefits of trials, that led to the use of that morphometry in approval of uh elendronate. A guy named uh Rob Epstein had used a similar kind of approach for the early studies of elendronate that that you know that contributed to its approval. He's an epidemiologist, not a radiologist, and used these kinds of tools. And so that kind of that morphometry allowed us to screen in women that we called having vertebral fractures, and also then to look at the endpoint of occurrence of vertebral fractures as one of the uh benefits of lendronate. So, yeah, yeah, having a definition of vertebral fractures that was feasible to assess in large numbers of patients and accepted by the community, I think, was was really an important step forward. It continues to evolve. Your arguments about which is best uh still go on, but with less intensity uh because uh it's change in the heights that matters.

SPEAKER_00 16:44

Well, as a geneticist, I find all of this extremely interesting because from a genetics point of view, we know the vertebral osteoporosis and hip osteoporosis do not have the same loci. So having these definitions that allow us to just not make generalizations based on the hip has been so instrumental in my career.

SPEAKER_01 17:06

So great. So Yep. I think one of the things that Soft done that we did in summary, really, is we show the value of rigorous epidemiology, rigorous designs of studies with large and adequate sample sizes, and the adoption of very innovative methods like vertebral fracture assessment, dual X-ray absorptiometry, state-of-the-art measurements of risk factors. And my, you know, if if if I have a recommendation to generations to come, is to pay attention to the rigor, sample size estimates that are adequate for things that are important, and adopting innovative tools, innovative ways of measuring, innovative ways of thinking about the condition of osteoporosis. So, I mean, I think those are the pillars on which SOF and I think my career was based.

SPEAKER_00 17:57

I think that that's excellent advice for the future generation. I almost want to take that and put it as a plaque up in my lab and or in my lab meetings as we talk about or in our journal clubs as we talk about what does this study mean? How is this study impactful? How is your work impactful? So I think that's great advice.

SPEAKER_01 18:19

Um one last point is about the importance of this is the 40th anniversary about ASBMR and JBMR.

SPEAKER_00 18:26

Yes, it is.

SPEAKER_01 18:27

Well, that was a friendly place to publish some of our early work on bone density and fractures. A lot of it was published in the New England Journal and JAMA, but boy, the JBMR has been the home. But maybe more importantly, ASBMR became the place where people doing clinical research eventually began to come together. There was no other network, there was no other place. And so the gathering and the interactions at ASBMR from those who were then publishing in JBMR was critical to advancing the field and engaging companies, companies who would attend those meetings and provide opportunities for additional studies of drugs. So yay, ASBMR!

SPEAKER_00 19:11

I agree. Yes, it's gonna be a very fabulous celebration. I hope to see you in Boston. Uh, Dr. Cummings, it's been such an amazing pleasure to have this interview with you and get to hear some of these stories. I'm so glad that I'm actually going to take some of these recordings and deploy them for my grad students well before the podcast comes about. So thank you so much for taking this time. This is wonderful.

SPEAKER_01 19:40

Well, thank you. Thanks for the opportunity to tell them a bit about the history that people tend to forget.

SPEAKER_00 19:45

Well, I think that the history is important because if you don't understand what a study is done or then a study has done, how do you design the next steps forward? So excellent advice and excellent um storytelling. So thank you.

SPEAKER_01 20:01

Thank you.