The study published August 29 in Nature Communications shows that artificial intelligence and machine learning technologies can effectively and efficiently subtype pathology samples from patients with RA.

For this study, Dr. Richard Bell, an Instructor in the HSS Research Institute and Arthritis and Tissue Degeneration Program, and Lionel B. Ivashkiv,  MD, Chief Scientific Officer at HSS teamed up with  Dr. Fei Wang, a professor of population health sciences and the founding director of the Institute of AI for Digital Health (AIDH) in the Department of Population Health Sciences at Weill Cornell Medicine. 

“Our study addresses the analytical bottleneck of pathology research,” Dr. Bell said. “It is very time-consuming and tedious.” “Our tool automates the analysis of pathology slides, which may one day lead to more precise and efficient disease diagnosis and personalized treatment for RA,” said Dr. Wang. “It shows that machine learning can potentially transform pathological assessment of many diseases.”

There are several existing machine learning tools for automatic analysis of pathology slides in oncology. Dr. Wang and his colleagues have been working to expand the use of this technology in other clinical specialties.

Digital Pathology and Precision Medicine in RA

Distinguishing between the three subtypes of RA may help clinicians choose which therapy is most likely to be effective for a particular patient. This personalized medicine approach would represent a breakthrough in the care of patients with RA, and is a major goal of the Research Institute and Division of Rheumatology at HSS, which involves multiple laboratory and clinical investigators. “It’s the first step towards more personalized RA care,” Dr. Bell said. “If you can build an algorithm that identifies a patient’s subtype, you’ll be able to get patients the treatments they need more quickly.”

 The technology may provide new insights into the disease by detecting unexpected tissue changes that humans might miss. By saving pathologists time on subtyping, the tool may also decrease the cost and increase the efficiency of clinical trials testing treatments for patients with different subtypes of RA. “This work represents an important advance in analyzing RA tissues that can be applied for the benefit of patients” Dr. Ivashkiv said.

 Pathologists currently manually classify arthritis subtypes using a rubric to identify cell and tissue characteristics in biopsy samples from human patients—a slow process that adds to the cost of research and may lead to inconsistencies between pathologists.

 The team first trained its algorithm on RA samples from one set of preclinical models, optimizing its ability to distinguish tissue and cell types in the sample and sort them by subtype. They validated the tool on a second set of samples. The tool also yielded new insights into treatment effects in the models, such as reduced cartilage degradation within six weeks of administering commonly used RA treatments.

Then, they deployed the tool on patient biopsy samples from the NIH-supported Accelerating Medicines Partnership Rheumatoid Arthritis research consortium, of which HSS is a major participating site in research led by Laura Donlin, PhD, scientist and co-director of the HSS Precision Medicine Program and rheumatologist Susan  M. Goodman, MD. The new digital pathology approach could effectively and efficiently type human clinical samples. The researchers are now validating the tool with additional patient samples and determining the best way to incorporate this new tool into pathologists’ workflows.

 The team is working to develop similar tools for evaluating osteoarthritis, disc degeneration and tendinopathy.

The research reported in this story was supported in part by the National Institute of Arthritis and Musculoskeletal and Skin Diseases, the National Institute of Allergy and Infectious Diseases, the National Institute of General Medical Sciences, and the National Institute on Aging, all part of the National Institutes of Health, through grant numbers R21AR071670, R01AI175212, UH2AR067690, UC2AR081025, F30AG076326, T32GM007356, AR046713, AR050401, R01AR078268, UC2AR081025, UL1TR001866 and R01AR056702.

The new single-blind, randomized, pilot, clinical trial, part of the Arthritis Foundation 2023 FastOA Interventional Pilot Studies, will harness the expertise of surgeons, radiologists, biostatisticians, and cellular and molecular biologists to determine whether cBMA may prevent the chronic inflammation that can lead to pain and stiffness following surgery, and can also contribute to post-traumatic osteoarthritis. Previous studies have suggested that cBMA injections may improve surgical outcomes, but this approach has not yet been evaluated in rigorous clinical trials in patients undergoing ACL revision surgery.

The cBMA used in these procedures is the patient’s own and is collected from the pelvic bones. The participating patients will be randomized to receive either cBMA injections at the time of revision surgery or standard care.

The goal of the trial is not only to establish whether the addition of cBMA reduces the risk of PTOA, but also to understand the underlying mechanisms of cBMA therapy and to look for biomarkers in the joint microenvironment that might help to refine the approach and further improve outcomes. Given the lack of effective non-surgical interventions for the prevention and treatment of PTOA, this study is expected to provide valuable insights that could guide care in the future.

To determine whether any differences observed between the two arms of the trial are clinically meaningful, the primary measure of the treatment’s success will be the patients' self-reported pain levels 12 months after surgery. This will be determined by using the Knee Injury and Osteoarthritis Outcome Score (KOOS) for pain. As a secondary measure, the investigators will deploy advanced, quantitative MRI scans to look for changes in the cartilage. “We will leverage clinical outcomes and imaging to evaluate the efficacy of adding cBMA to standard treatment,” said co-primary investigator Miguel Otero, PhD, scientist in the Orthopedic Soft Tissue Research Program, co-director of the Derfner Foundation Precision Medicine Laboratory, and of the HSS Research Institute Histopathology Service.

“Depending on what we learn about whether cBMA reduces the risk of PTOA, the results generated by this study may have the potential to change clinical practice,” added co-primary investigator Scott Rodeo, MD, Vice Chair of Orthopaedic Research, co-director of the Orthopaedic Soft Tissue Research Program, and director of the Center for Regenerative Medicine at HSS.

By analyzing the cellular and molecular make-up of the cBMA as well as synovial fluids collected during and after the ACL revision surgery, the investigators also hope to gain insight into the populations of immune cells present within the joint microenvironment and the link between certain immune components and clinical outcomes. The researchers hypothesize that cBMA’s functionally relevant immune-modulatory and disease-modifying effects depend on the fluid’s cellular and molecular composition. They expect that these biospecimens may yield clues about how to further improve patient outcomes.

The FastOA2023 trial will harness strength of two Arthritis Foundation OACTN Clinical Trial Units that are also participating in the Post-Injury Knee Arthritis Severity Outcomes (PIKASO) Study and will use parallel instruments and dataset management. PIKASO is an ongoing, multicenter, randomized phase II clinical trial examining the use of metformin in patients who are at high-risk for developing PTOA in the knee after ACL reconstruction.

To learn more about the study and when enrollment opens in FastOA2023 visit clinicaltrials.gov. 

Researchers at HSS and other institutions have noted that women with knee osteoarthritis report more pain than men, but the reason for this difference has been unclear. “Others have speculated that women tend to delay surgery more than men, but when we looked in our database, that wasn’t true,” said HSS rheumatologist Dana Orange, MD, MS, assistant professor at The Rockefeller University and senior author of the study. “We studied synovial tissue removed at the time of total knee arthroplasty to look for a biological reason that may explain the difference in reported pain between sexes.”  

Synovial tissue lines the knee joint and produces fluid that helps the joint move. It can become inflamed as osteoarthritis progresses. Pathologists at HSS have been generating data on different types of cells found in synovial tissue, including mast cells, which are also found in many other tissues throughout the body and are commonly known for producing inflammatory chemicals called histamines during allergic reactions and asthma. Basic research by other scientists has suggested a link between mast cells and osteoarthritis progression and pain.

Dr. Orange and colleagues, including HSS rheumatologist Bella Mehta, MBBS, MS, lead author of the study, studied joint tissue obtained from 96 women and 61 men who underwent total knee replacement at HSS. They counted the number of cells for more than a dozen cell types typically found in synovial tissue and examined synovial fluid and blood using a high-powered microscope. They also evaluated patient-reported pain outcomes collected with two validated surveys. “At HSS, we were in a unique position to conduct this research because we have synovial tissues collected from patients with end-stage osteoarthritis undergoing arthroplasty and expert musculoskeletal pathologists who systematically grade tissue samples for 13 characteristics,” Dr. Mehta said.  

The investigators discovered that synovial tissue from women had significantly more mast cells, 63 per sample area, compared to 46 in tissue from men, on average. They also found higher levels of a byproduct of mast cells called tryptase in synovial tissue from women than men, providing further evidence of increased mast cell activity. No other differences in synovial tissue between sexes were observed. Finally, as expected, women reported worse pain than men on both surveys.

“We hope our findings encourage other researchers to start thinking about biological factors that may contribute to sex differences in patient-reported pain in knee osteoarthritis,” Dr. Orange said.

Authors: Bella Mehta, MBBS, MS, Edward F. Dicarlo, MD, Miguel Otero, PhD, Peter K. Sculco, MD, Mark P. Figgie, MD, Susan M. Goodman, MD (HSS), Fei Wang, PhD (Weill Cornell Medicine), Eun Kyung Song, PhD (Stanford University School of Medicine), Dana Orange, MD, MS (HSS and The Rockefeller University). 

TKR is often the only management option for patients with severe knee joint damage. Identifying which disease caused the joint damage is essential for guiding treatment plans, given that RA is a systemic, inflammatory disease that may also affect the eyes or lining around the heart, while OA affects just the joints. “We know there are many more immune cells present in the synovium, or joint tissue, of patients with RA compared to those with OA,” said Bella Mehta, MBBS, MS, rheumatologist at HSS and lead author of the study. “But precisely how many more has not been clear.”

“Pathologists typically assess images of synovium to determine the extent of inflammation using a combination of approaches, including assigning the level of immune cell infiltration on a scale from 0 to 4,” said Dana Orange, MD, MS, rheumatologist at HSS, assistant professor at Rockefeller University and senior author of the study. “However, these methods are imperfect.” For example, a recent study by HSS investigators found that assessments from two highly experienced pathologists evaluating the infiltration of one type of immune cells known as lymphocytes on the same slides agreed only 67 percent of the time.¹

Drs. Orange, Mehta and colleagues at HSS and collaborating institutions developed and validated a computer vision tool that rapidly counts tens of thousands of cell nuclei in whole-slide images of synovium.² For their present study, they measured 14 different pathologist-scored features in synovium from 60 patients with RA and 147 patients with OA who underwent TKR, and used the computer vision tool to determine cell density.

The investigators identified significant differences between RA and OA features in synovium. The RA samples showed increased cell density; low numbers of mast cells, a type of white blood cell; and lower evidence of fibrosis or scarring compared to the OA samples. The probability of correctly distinguishing between RA and OA in synovium was 85 percent when using the 14 pathologist-scored features alone, 88 percent when using the computer’s score for cell density alone and 91 percent when the researchers combined the pathologists’ scores and the computer’s cell density calculation. The researchers determined a cutoff point for distinguishing RA from OA, determining that synovium containing more than 3,400 cells per mm² should be classified as RA.

“While our innovation is not ready for clinical use yet, it holds promise for assisting pathologists in the future,” Dr. Orange said. “Right now, we see it as a valuable tool for research purposes because it provides an accurate and 100% reproducible score of inflammation and look forward to developing it further.”

Dr. Orange added that in the future computer vision could be trained to glean other types of information from tissue samples, including which types of cells are present and whether they are close enough together that they are likely to be communicating with each other. This more granular assessment might enable clinicians to know more precisely which cells are causing tissue damage and tailor treatments accordingly.

Authors: Bella Mehta, MBBS, MS, Susan M. Goodman, MD, Edward F. DiCarlo, MD, Deanna Jannat-Khah, J. Alex Gibbons, Miguel Otero, PhD, Laura Donlin, PhD (HSS), Tania Pannellini, MD, PhD (Weill Cornell Medicine), William Robinson, MD, PhD (Stanford University), Peter K. Sculco, MD, Mark P. Figgie, MD, Jose A. Rodriguez, MD (HSS), Jessica Kirschmann (Stanford University), James Thompson, David Slater, Damon Frezza (The MITRE Corporation), Zhenxing Xu, Fei Wang, PhD (Weill Cornell Medicine), Dana Orange, MD, MS (HSS and Rockefeller University).

References

1. Orange DE, Agius P, DiCarlo EF, et al. Identification of Three Rheumatoid Arthritis Disease Subtypes by Machine Learning Integration of Synovial Histologic Features and RNA Sequencing Data. Arthritis Rheumatol. 2018;70(5):690-701. doi:10.1002/art.40428

2. Guan S, Mehta B, Slater D, et al. Rheumatoid Arthritis Synovial Inflammation Quantification Using Computer Vision. ACR Open Rheumatol. 2022;4(4):322-331. doi:10.1002/acr2.11381

HSS is the world’s largest academic medical center specialized in musculoskeletal health, spanning orthopedics, rheumatology and related disciplines. The HSS Research Institute maintains 20 laboratories dedicated to solving debilitating orthopedic and rheumatic conditions such as arthritis, bone and soft tissue injuries, autoimmune diseases, and musculoskeletal pain and deformities. There, more than 300 dedicated personnel focus on tissue repair, improving surgical outcomes, autoimmunity and inflammation, genomics, new treatments, and precision medicine.

“HSS has a long history of contributing to the collective base of clinical and basic science knowledge and finding healthcare solutions for complex conditions,” said Louis A. Shapiro, President and CEO, HSS. “We’re proud that through the joint efforts of our institution and philanthropic support, we will have the ability to make a strong impact on this growing and vital area of research.”

“As experts in inflammatory disorders and in the development of interventions for overactive immune responses, the clinicians and researchers at HSS are well-positioned to investigate many of the adverse effects of COVID-19,” says Lionel B. Ivashkiv, MD, Chief Scientific Officer at the HSS Research Institute. “This includes studying the causes of these adverse effects as well as how to prevent and treat them.”What follows are descriptions of the first group of funded projects in basic/translational research:

Activation of pDCs by SARS-CoV-2 and Its Impact on Macrophage Response

Principal Investigator: Franck Barrat, PhD

Co-Investigator: Marie-Dominique Ah Kioon, PhD

This project will study cell types that are responsible for cytokine storm syndrome — the hyperactive immune response seen in people with COVID-19 — by looking at how certain immune cells are activated by SARS-CoV-2. Research in mice infected with SARS-CoV, a coronavirus similar to the one that causes COVID-19, has suggested that plasmacytoid dendritic cell precursors (pDCs) are key to the immune response to infection. These pDCs activate macrophages, which in turn secrete cytokines. In the SARS-CoV research, depletion of pDCs appeared to protect the mice from lethal lung injury. Using blood samples from donors, the investigators will study the pathway by which pDCs activate macrophages and look at ways to therapeutically block that process.

Inhibiting RNA Polymerase II Transcription Complexes in Macrophages to Target COVID-19–Associated Cytokine Storm

Principal Investigator: Inez Rogatsky, PhD

Co-Investigators: Steven Josefowicz, PhD, and Robert P. Fisher, MD, PhD

This pilot project will dissect the role of macrophages in SARS-CoV-2-induced acute respiratory distress syndrome (ARDS), the main driver of COVID-19-associated mortality. We will test small-molecule inhibitors of RNA Polymerase II (Pol II) transcription complexes for their ability to modulate type I interferon and inflammatory pathways in monocytes/macrophages. This research will be done using cultured macrophages as well as donor blood and blood from COVID-19 patients.

Mechanisms of Cytokine Storm in Patients with COVID-19

Principal Investigator: Mary K. Crow, MD

Co-Investigators: Mikhail Olferiev, MD, and Lionel B. Ivashkiv, MD

The objectives of this study are to describe the process of the cytokine storm in people with COVID-19 and to identify biologic predictors of a favorable outcome in patients with severe cases of the disease. The project aims to characterize immune cell populations seen in COVID-19 patients who experience cytokine storm and compare them to those patients who do not, to compare the immune response before and after patients are given the anti-inflammatory drug anakinra, and to identify measures that suggest patients are more likely to decline and eventually require mechanical ventilation. The research will employ blood samples from HSS patients who are being treated for COVID-19 at New York–Presbyterian Hospital and who meet certain other qualifications.

What follows are descriptions of the first group of funded projects in the areas of clinical and health outcomes research:

Response to and Recovery from TKA in Patients with Antibodies to SARS-CoV-2

Principal Investigator: Miguel Otero, PhD

Co-Investigators: Meghan Kirksey, MD, PhD, and Peter K. Sculco, MD

It is unknown if people who have been exposed to COVID-19 may be at higher risk of experiencing an abnormal immune response following surgery, resulting in poor outcomes. This study will evaluate the response to and recovery from total knee arthroplasty (TKA) in people who have antibodies to COVID-19 — a marker of exposure. This study will include both patients who have COVID-19 antibodies and those who don’t, to act as controls. Patients will be followed for six weeks after surgery and evaluated for the presence of certain immune markers in the blood, as well as symptoms of inflammation including pain and stiffness in the joint.

SARS-CoV-2 Exposure and the Role of Vitamin D Among Hospital Employees

Principal Investigator: Emily M. Stein, MD, MS

Co-Investigators: Theresa T. Lu, MD, PhD; Andy O. Miller, MD; Jeri Nieves, PhD; and Alana Serota, MD

It is unknown if people with vitamin D deficiency may be more likely to become infected with COVID-19. This study will investigate vitamin D status and associated immune markers as risk factors for COVID-19 infection in a cohort of healthcare workers. Healthcare workers are at higher risk of contracting COVID-19 than the general population, making them a good group to study. Vitamin D is critical for immune function and is known to be protective against respiratory-tract infection and tuberculosis. This prospective, observational study will follow healthcare workers at HSS and at other healthcare facilities for one year, to determine whether levels of vitamin D and certain immune cells in the blood make someone more susceptible to COVID-19 infection.

Association of Immunomodulatory Medication Use and Social Determinants of Health with COVID-19 Infection in Systemic Rheumatic Disease Patients in New York City

Principal Investigator: Medha Barbhaiya, MD, MPH

Co-Investigators from HSS: Lisa Mandl, MD, MPH; Catherine MacLean, MD, PhD; Vinicius Antao, MD, PhD; Jane Salmon, MD; and Mayu Sasaki, MPH

Other Co-Investigators: Candace Feldman, MD, MPH (of Brigham and Women’s Hospital); Debra D’Angelo, MS (of Weill Cornell Medicine)

Using data from the INSIGHT Clinical Research Network, a central repository containing longitudinal electronic health data for residents of New York City, investigators will assemble a cohort of patients being treated with immunomodulatory medications for rheumatic disease. This patient population will then be used to study the effect of these medications on COVID-19 incidence and outcomes. Retrospective data will be used to evaluate the incidence and severity of COVID-19 in these patients. Patients will also be studied prospectively to determine whether there’s a relationship between COVID-19 infection and future rheumatic disease as well as to study connections between infection and future psycho-social issues.

Assessment of Surgical Outcomes in the COVID-19 Pandemic Era

Principal Investigator: Andy O. Miller, MD

Co-Investigators: Scott A. Rodeo, MD, and Mark Fontana, PhD

Investigators will implement a patient registry to evaluate how COVID-19 affects outcomes and complication rates after orthopedic surgery. This registry, along with COVID-19 screening procedures, will provide the tools to address specific research questions. Among these questions are determining the incidence of current and prior infection among the HSS surgical population, the clinical features associated with current and prior infections in this patient population, and whether COVID-19 status affects short-term complication rates.

What follows is a description of an integrated multidisciplinary study being undertaken jointly by the Adult Reconstruction and Joint Replacement (ARJR) Perioperative Research Group, Anesthesiology and Rheumatology:

Prediction and Prevention of Postoperative Blood Clots in COVID-19 Patients

Principal Investigators: Friedrich Boettner, MD; Kethy M. Jules-Elysee, MD; Lisa A. Mandl, MD, MPH

Co-Investigators: ARJR surgeons: Alejandro Gonzalez Della Valle, MD; Jason Blevins, MD; David J. Mayman, MD; Peter K. Sculco, MD; Geoffrey H. Westrich, MD and Thomas P. Sculco, MD

Medicine/Rheumatology: Medha Barbhaiya, MD, MPH; Doruk Erkan, MD, MPH; Deanna Jannat-Khah, DrPH

Pathology: Thomas W. Bauer, MD, PhD

Anesthesiology: Stavros G. Memtsoudis, MD, PhD, MBA; Alexandra Sideris, PhD

ARJR: Amethia Joseph, MHA; Ethan Krell, MS

Weill Cornell Medicine: Raymond David Pastore, MD

Recent literature suggests that one of the major complications seen in people with COVID-19 is thrombosis (the formation of blood clots) due to endothelial dysfunction, persistent inflammation and potentially antiphospholipid antibodies. As elective surgeries resume, those with prior exposure to SARS-CoV-2 will inevitably present for treatment, and some may have perioperative management considerations related to their risk of deep-vein thrombosis. This project will use, a noninvasive device that can determine clotting risks, to investigate whether people who have had COVID-19 have a more dysfunctional endothelium preoperatively and at 24 hours after surgery. The investigators will measure antiphospholipid antibodies and inflammatory markers, and evaluate the prevalence of asymptomatic post-operative deep-vein thrombosis in people who undergo TKR and have SARS-CoV-2 antibodies.

COVID-19 Translational Research Core at HSS

Principal Investigator: Theresa Lu, MD, PhD

Co-Investigators: Jessica Andrés-Bergós, PhD; Miguel Otero, PhD and Emily M. Stein, MD, MS

The COVID-19 Translational Research Core (TRC) was designed to fill critical gaps in the resources needed to promote the broad range of COVID-19–related clinical and translational research at HSS. The TRC will provide consultation on the design and implementation of COVID-19 research in the areas of orthopedics, rheumatology and metabolic bone disease; support for a COVID-19 biobank; and technical expertise and facilities required for clinical and translational researchers working on COVID-19–related projects. The TRC staff will help to acquire, house, and track biospecimens from investigator-initiated COVID-19–related research studies.

Knee replacement surgery, or total knee arthroplasty (TKA), is one of the most common orthopedic operations, with more than 600,000 performed annually in the United States. Although TKA is a very successful procedure, up to 20% of patients report dissatisfaction after surgery, including debilitating stiffness and pain that may eventually lead to a second (revision) surgery.

“In some cases we can identify potential reasons for the development of stiffness, but a subset of patients develop stiffness and pain for reasons that we can’t identify,” said Miguel Otero, PhD, assistant scientist in the Orthopedic Soft Tissue Research Program and co-director of the Derfner Foundation Precision Medicine Laboratory at HSS. These patients are said to have arthrofibrosis, a condition marked by excessive formation of scar-like tissue that is associated with restricted movement in the joint and pain in the surgically repaired knee.

For the study titled “RNAseq Analyses of Neo-Synovial Tissues Retrieved at the Time of Revision Surgery from Patients with Stiff Knees Uncovered Arthrofibrosis-Specific Gene Signatures,” Dr. Otero and his colleagues integrated clinical outcomes of TKA patients with imaging studies and genomic information to see if they could identify unique gene signatures in patients who developed arthrofibrosis.

They enrolled 80 HSS patients undergoing revision knee surgery after experiencing complications following TKA. The researchers separated patients with stiffness into two groups based on whether the problem was linked to a known reason, such as a mechanical issue, or an unknown cause (arthrofibrotic group). A third group, including patients who underwent revision not accompanied by stiffness, for complications including instability or aseptic loosening, served as control.

Dr. Otero’s team analyzed tissue samples taken from the joints of patients during their revision surgery to see if they could find changes in gene expression between the different groups. They identified 435 genes differentially expressed in tissues from patients with stiffness for identifiable reasons relative to the control group. Importantly, they found more pronounced differences (1,509 differentially expressed genes) comparing the arthrofibrosis and non-stiff groups.

“Think about surgery as a massive injury. The body needs to heal the damage, and scars are often a consequence of that healing process,” Dr. Otero said. “The excessive scarring and fibrosis that we see in patients with arthrofibrosis is an overreaction to that injury, but we don’t yet know why they are overreacting or why their reaction is different from other patients who also developed stiffness after TKA.”

Dr. Otero and his colleagues hope to winnow their large pool of genes to a much smaller subset that will serve as a more precise signature of arthrofibrosis after TKA. “We want to identify cellular and molecular signatures with therapeutic and prognostic value, so that we can develop preventative therapies targeting specific genes or pathways, and also identify patients who are at risk of developing arthrofibrosis following TKA,” he said.

One goal of the research, he said, is to find targeted and specific ways to prevent the development of arthrofibrosis in patients undergoing TKA. “Even though all patients with stiffness have scarring and limited joint motion, patients with arthrofibrosis do not necessarily behave the same as other patients with stiffness,” Dr. Otero said. “The hope is to implement precision medicine approaches so that, potentially, we can match at-risk patients with specific treatment options and prevent this debilitating problem.”