Using New Autoantibodies in Rheumatic Disease: An Update

Kevin D. Deane, MD, PhD


May 23, 2019

Antibody testing is likely to play an increasingly important role in the practice of rheumatology. The classification of lupus is enabled by antinuclear antibodies, as is rheumatoid arthritis (RA), by employing autoantibodies such as rheumatoid factor (RF) and antibodies to citrullinated protein/peptide antigens (ACPAs).

Evidence suggests that autoantibodies can help determine disease manifestations, prognosis, and even comorbidities in rheumatic conditions. For example, a high level of double-stranded DNA is associated with increased risk for renal disease in lupus. In addition, high levels of RF and ACPAs are associated with a more persistent and damaging course of RA.[1]

Furthermore, a number of autoantibodies in idiopathic inflammatory myopathies, such as dermatomyositis and polymyositis, have been associated with severity of disease and even the presence of concomitant cancer.[2,3] Specifically, antibodies to Mi-2 are associated with less severe myositis, and antibodies to transcriptional intermediary factor 1 gamma (anti-TIF1-γ) have been associated with underlying cancer.[2,3]

Newer Autoantibodies

The testing that is currently clinically available in RA is typically for antibodies to cyclic citrullinated peptide/protein (anti-CCP). These are antibodies to proteins/peptides containing citrulline, which is the posttranslated modification of arginine. Anti-CCPs are highly specific for RA (typically > 90%) and aid in diagnostic certainty.[4] However, these assays utilize a proprietary set of antigens that have been created to optimize test performance. Therefore, clinicians do not know the specific antigen to which these autoantibodies are reacting.[5] The tests are still highly useful in determining a diagnosis and prognosis in RA. However, there is a growing understanding that identifying autoantibodies to specific citrullinated proteins/peptides may improve diagnosis and potentially help determine prognosis.

Schwenzer and colleagues[6] have found that testing for certain antibodies to the proteins citrullinated alpha-enolase peptide-1 (anti-CEP-1) and tenascin-C (cTNC5) improved the sensitivity for RA over a standard anti-CCP assay (anti-CCP2). In particular, adding testing for anti-CEP-1 and anti-cTNC5 to anti-CCP2 increased sensitivity by a range of ~2.5% to 5% across several case-control cohorts, although specificity was not reported. In addition, patients who were positive for anti-CEP-1 and anti-cTNC5, as well as anti-CCP2, had more severe disease compared with those who were positive only for the latter. Although these results will need to be validated, overall they suggest that adding specific ACPAs to commercial tests such as anti-CCP2 may increase sensitivity and help identify a subset of subjects who may have more severe disease.

In the idiopathic inflammatory myopathies, and in particular dermatomyositis, certain autoantibodies are associated with cancer. Anti-TIF1-γ has been strongly associated with increased risk for underlying cancer. However, studies have also shown that not all individuals with elevations of this autoantibody have cancer.[7]

To explore this further, Aussy and colleagues[8] evaluated 51 individuals with dermatomyositis, all of whom had anti-TIF1-γ positivity, as identified using standard assays. They found that of these 51 individuals, 40 (78%) had cancer that included solid tumors (~88%) and hematologic malignancies (~13%).

All individuals were positive for the antibody; however, within this group, additional positivity for the isotype IgG2 anti-TIF1-γ was strongly associated with cancer. Furthermore, levels of IgG2 anti-TIF1-γ greater than 385 units had a 100% positive predictive value for cancer. In addition, elevations were associated with increased risk for overall mortality. The authors concluded that the IgG2 isotype of anti-TIF1-γ could add new information regarding the prediction of cancer risk and overall mortality.

Scleroderma's Association With Cancer

Scleroderma also has an association with underlying cancer, and in particular elevations of antibodies to RNA polymerase III large subunit (anti-RPC155). Similar to what has been seen in dermatomyositis with anti-TIF1-γ antibodies, however, not all anti-RPC155-positive individuals with scleroderma have an underlying cancer identified.[9,10]

To investigate this issue, Shah and colleagues[11] studied a cohort of 168 individuals with scleroderma and anti-RPC155 positivity, 80 with cancer and 88 without. They performed antibody discovery using immunoprecipitation methods, and found that the subset of anti-RPC155 positive individuals with scleroderma but without cancer were significantly more likely to also have antibodies to the RNA polymerase I large subunit (anti-RPA194): ~18% in those without cancer vs ~4% in those with cancer (p=0.003). The authors concluded that the immune responses evidenced by the anti-RPA194 antibody may play a role in influencing cancer emergence, perhaps through influencing RNA polymerase, which is known to have increased activity in some cancers.[12] They also concluded that testing for several autoantibodies may help refine cancer risk stratification.

Thoughts on Autoantibodies and Patient Care Management

There is a growing understanding that autoantibodies can be used in clinical care in the diagnosis, prognosis, and identification of associated diseases such as cancer. However, expanding the use of these autoantibodies presents unique challenges, a major one being the question of how to readily interpret these findings to guide clinical care.

In terms of improving diagnostic criteria for a disease, it may be more straightforward. For example, if new biomarkers improve the sensitivity and specificity for a disease over their older counterparts, then diagnostic certainty can be improved, possibly along with subsequent management. Specifically, when faced with an individual with new inflammatory arthritis, finding an RA-specific antibody helps with diagnostic certainty and may in turn lead to faster initiation of appropriate therapy, which could otherwise be delayed while additional evaluation is performed to identify a classification for the clinical findings. It can also reduce concerns about some other potentially more dangerous condition that could present with similar clinical findings. For example, the presence of anti-Mi2 in myositis can help determine that this is an autoimmune muscle disease rather than a toxic metabolic process or genetic process causing the myopathy.

It is less clear how to proceed when autoantibodies are associated with disease severity. For example, if an autoantibody is positive and associated with severe disease in RA, how should that guide therapy, which is typically driven by treating the clinical manifestations of disease? If an individual has autoantibodies associated with severe RA, but is doing well with current therapy, should management be changed? Probably not.

Conversely, it is difficult to know how to use autoantibodies that may be associated with less severe disease. This may be the case with the anti-Mi2 antibody. If an individual has myositis and this autoantibody, and is doing well, that is fairly straightforward. However, if they have myositis and this autoantibody, but are doing poorly, treatment will have to be based on the clinical condition, not just the autoantibody.

Management decisions are particularly difficult when an autoantibody is associated with a comorbidity such as cancer. In particular, if the anti-X has been associated with cancer, and that antibody is elevated in an individual with disease Y, how should one look for cancer? Should a standard age-appropriate workup that may include mammograms, colonoscopy, cervical evaluation be used? Or, should more detailed evaluations be done, looking for harder-to-find cancers like ovarian cancer, or pancreatic cancer using advanced imaging (eg, MRI, PET) and potentially blood-based biomarkers for cancer such as CA-125 or cancer-associated DNA?[13] Guidelines on how to evaluate for myositis-associated cancer are forthcoming, and this should help standardize and optimize approaches.

Conversely, it is difficult to know how to use the presence of an autoantibody that may be associated with 'protection' from cancer, particularly if that protection is not 100%. For example, Shah and colleagues[11] found that the anti-RPA194 antibody was more common in individuals without cancer, but it still was present in several individuals with cancer. Additional studies are needed to help establish exactly how these tests can be used in clinical care.

The biology of associations between autoantibodies and disease presence and severity, or other outcomes such as cancer, is fascinating. In particular, the association of autoantibodies to specific targets and decreased risk for cancer may help lead to breakthroughs in how to treat or prevent it. Indeed, as mentioned in the article by Shah and colleagues, it may be that anti-RPA194 reduces its target RNA polymerase I activity, which in turn protects against cancer—a model that is suggested by direct RNA polymerase inhibitors showing some benefit in animal models of cancer[12] and anti-DNA antibodies leading to the death of cancer cells.[14] Hopefully, further studies in these areas will ultimately help us better understand disease pathogenesis and lead to improved therapies or preventive interventions for disease.

Finally, although not all of the autoantibodies discussed here are widely available in routine clinical practice, many are still available nonetheless, with more potentially accessible in approved forms for clinical use in the near future. This will result in a growing number of autoantibodies across a variety of diseases, which in turn presents clinicians with new issues on how to manage the information, make good clinical decisions, and impart that information back to individuals with disease. This is especially fraught given the wide access that individuals with disease have to their lab test results and information about those results from the Internet. For example, one can readily imagine someone with myositis and anti-TIF1-γ antibody positivity finding on the Internet that the latter is associated with cancer, and having significant distress until they can talk through the meaning of the test and workup with their clinician.

One way forward in this area would be for clinicians to partner with laboratory testing agencies to develop clear algorithms to use these expanding autoantibodies in practice. Another is to develop clear educational materials that can help individuals with disease understand their test results, optimizing the relationships between healthcare providers and individuals with disease.

Follow Medscape on Facebook, Twitter, Instagram, and YouTube


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.
Post as: