Read Part I of the series here.

Andrew L. Concoff, MD, FACR, CAQSM

Part II: Systemic Lupus Erythematosus (SLE)

First, the good news: tremendous progress has been made toward revolutionizing the evaluation and treatment of lupus nephritis (LN). By applying analysis of urine proteomics, a commercialized urine liquid biopsy may soon be developed to replace renal biopsy for the identification of lupus nephritis (LN). Similarly, substantial optimism has been generated by a veritable avalanche of 4 new medications that have demonstrated favorable results in the treatment of LN. Of the 4, 2 have been approved, 1 is undergoing the dreaded Phase III trial (which until recently had been where so many promising new therapies had gone to die), and 1 has completed Phase II with report due out very soon.

However, not all of the recent developments in SLE have painted as rosy a picture of the future. A great deal of attention has been brought to recent efforts to borrow the treat-to-target (T2T) approach from other chronic diseases, notably rheumatoid arthritis (RA), for application to the real-world management of SLE. Yet several barriers remain that may call into question whether T2T in SLE is ready for prime time. Finally, patient nonadherence, the dirty little secret of the practice of Rheumatology, and the resulting negative impact on outcome, has returned to the spotlight in light of recent studies. In particular, the tremendous magnitude of the increased mortality seen among hydroxychloroquine (HCQ) nonadherent patients in SLE is a stark reminder of the critical importance of rheumatologists learning to better assess and optimize adherence with treatment, an obligation that is often woefully unmet.

With the landmark study of 2020¹ in SLE, the era of meta-omics analysis has arrived in SLE. This revolution in cutting-edge, clinically relevant, basic science research is largely due to the substantial investment of time, effort, and dollars in the Accelerating Medicines Partnership (AMP).² For those Rheumatologists, particularly clinicians, who have not kept up with the progress of the AMP researchers in SLE and RA, the time is now to do so. The initial fruits of this labor provide clear hints as to the future of clinical care in these diseases and, through derivative follow, other rheumatic diseases. The latest AMP in SLE study¹ demonstrates the potential value of urine proteomics paired with single-cell genomics that may portend replacement of renal biopsy for LN with the development of urine-based, liquid biopsies. An elegant series of urine proteomics studies tracking 1000 urine protein biomarkers identified an immune activation gradient in active LN. By comparing with subsequent renal single-cell RNA-Seq results, it was recognized that this signature is driven predominantly by interferon (IFN)-γ from infiltrating immune cells. As the authors correctly describe in introducing the topic, the promise of these findings may go far beyond the obvious reduction in the invasiveness of the diagnostic testing required to evaluate LN.

Numerous, recognized limitations of the current, histopathologically-based treatment paradigm may be avoided by the proposed switch to liquid biopsy. Specifically, transitions in histopathologic class have been noted in prior repeat biopsy studies. Histopathological transitions within individual patients speak to a dynamic process that is neither reflected in the current single-biopsy-based paradigm nor easily addressed, given the invasiveness of the procedure, by sequential biopsy in the real-world. The development of a non-invasive test would be expected to result in more aggressive testing of milder, less clear-cut cases where biopsy may be delayed because of its invasiveness. Similarly, the lack of invasiveness would encourage more frequent repeat testing. In both cases, a urine-based, liquid biopsy would allow recognition of active nephritis at an earlier stage and, therefore, improve outcomes.

Additional factors favor urine-based liquid biopsy over the conventional approach. Interstitial renal disease is recognized to have pathophysiologic relevance but is not adequately accounted for by current histopathologic scoring. Renal biopsy results may be confounded by sampling error, a factor not seen in liquid biopsy. These and other factors likely contribute to variable outcomes that have been demonstrated among patients within a given histologic class; variability that may be minimized by an effective urine-based, liquid biopsy. Even more importantly, histopathologic scoring in LN appears to have limited ability to serve as a companion diagnostic test to predict individual-level medication response, in much the same manner as synovial biopsy recently was in RA.³ Histopathology simply cannot distinguish individual-level differences in the underlying molecular pathways that give rise to the shared phenotypes demonstrated at biopsy. By contrast, recent studies^1,4-6 have identified urine biomarkers, including IL-16 and TGFb1, that correlate with initial histopathology but that also decrease among responders to therapy (e.g., IL-16, TGFb1, and CD163). Though significant work remains to validate the findings of these studies and generate from them a commercially available, urine-derived liquid biopsy, the path forward in the diagnosis of LN will almost certainly leverage this approach in the future.

Advances in the diagnosis of LN seem to be arriving just in time, as numerous new therapeutic options are now, or soon will be, available to treat it. In 2011, when belimumab was approved, it marked the only approved biologic Disease-modifying antirheumatic drug (DMARD) for SLE, and the only newly approved SLE drug in nearly 50 years. In December 2020, with data from a 2-year, randomized controlled trial (RCT) demonstrating efficacy of belimumab plus standard of care besting placebo plus standard of care,⁷ belimumab also became the first approved treatment for adult LN. However, after voclosporin plus standard of care was demonstrated in a Phase III trial to be nearly twice as effective as the standard of care alone in delivering a renal response at 52 weeks,⁸ voclosporin became the second drug approved for adult LN just 1 month later. Next, obinutuzumab which had been approved to treat CLL in 2013 and was more recently approved for lymphoma, has been repurposed for the treatment of LN. After promising Phase II results at up to 2 years,⁹ a Phase III study of obinutuzumab in LN is currently enrolling¹⁰ with preliminary results expected in 2023. Finally, despite the disappointment of the prior TULIP-1 trial¹¹ which failed to meet its primary endpoint in treating extrarenal SLE, the reconfigured, BILAG-based Composite Lupus Assessment (BICLA)-based analysis of the TULIP-2 trial¹² demonstrated the superiority of anifrolumab over placebo. Filing for FDA approval of anifrolumab for extrarenal disease followed in 2020 and remains pending to date.But is anifrolumab effective in LN? The answers will soon be revealed as the Phase II, TULIP-LN1 trial has been completed¹³ but the results have not yet been released.

With numerous therapeutic options, the question becomes how is a rheumatologist to decide which of these new medications to choose to treat individuals with LN? Although, on the one hand, this is certainly a nice problem to have as prescribers will likely soon have to attempt to choose from amongst these medications. Furthermore, payers will have to determine which of these drugs to add to their formularies and in what relative position. It is extremely unlikely that head-to-head trials will be conducted. Thus, there is considerable risk that prescribing patterns could be driven by factors other than those conforming to the principles of precision medicine, determining the right drug, for the right patient, in the right dose, at the right time. Rather, devolution to the scenario seen in RA, where formulary financial arrangements rather than precision phenotyping and personal pathophysiology have driven medication choice, is possible if not likely. How might we be rescued from another disease riddled by the opaque and Byzantine system of monetary handoffs among a cavalcade of middlemen that have conspired to generate runaway costs of RA treatment? In the best case, companion diagnostic tests will be developed to allow prediction of response to these new drugs for LN at the individual level.As currently applied in oncology, only those with a high likelihood of response according to a medication-specific biomarker test would receive that treatment. Certain molecular pathways in SLE that are relevant for the current crop of new medications may be developed to serve this purpose. For instance, advances in the evaluation of differences in the interferon signature of individuals with SLE¹⁴ may identify those more likely to respond to IFN-directed therapies such as anifrolumab.¹⁵ The challenge is that no such biomarkers that pertain to these new medications for LN have yet been identified and the medications are likely to come to market in the absence of them. The search for companion biomarkers of response is planned in the next phase of the AMP SLE investigations.

A short-term solution in the absence of such biomarkers is the development of value-based drug pricing arrangements for the new SLE drugs. Under this model, a pharmaceutical manufacturer “bets” on the effectiveness of its medication by receiving full payment for the drug only when individual patients demonstrate a predetermined level of response by a specific timepoint. For those individual patients who do not meet response criteria, an outcome adjustment is made in the form of a payment from the manufacturer back to the payer, effectively lowering the overall cost of the drug. Which payer(s) might collaborate in such programs to assert control over cost of drug by linking formulary position to a manufacturer’s willingness to tie the drug cost directly to the value they deliver to individual patients? Which pharmaceutical manufacturer believes in its product enough to bet on the outcomes it will generate? Can such deals be brokered for these new medications? Only time will tell the answers to these questions. An addition to these arrangements that greatly enhances the likelihood of their success is the inclusion of a neutral, third-party to function as a fulcrum between the competing interests of payers and manufacturers in the 0-sum game of a value-based pricing arrangement. Taking a step back, however, opinions are mixed as to whether lupus researchers have identified and validated 1 further element that is required for both T2T approaches and value-based pricing arrangements: an adequately-validated, clinically applicable, and effective measure of disease activity.

A variety of attributes of SLE as a disease state make it a uniquely complicated challenge for the application of T2T principles or value-based contracts. The challenges are elucidated by comparing the attributes of SLE to RA, a disease in which both of these outcome-focused approaches have been applied to significant success. First, although RA can affect multiple organ systems, outcome measures that assess a single target organ system¹⁶ through joint counts along with physician and patient global assessments have been demonstrated to capture the risk of disease progression¹⁷ though not perhaps the entirety of the experience of disease from a patient perspective.¹⁸ In SLE, however, Protean manifestations across multiple organ systems challenge either the completeness with which an outcome measure captures the different domains on the one hand, or the applicability to the time constraints of real-world practice settings on the other. For example, according to one recent disease activity measure, the Lupus Low Disease Activity State (LLDAS),¹⁹ a patient may be considered to be in low disease activity despite having either significant arthritis or 2 of 3 other common flare manifestations (rash, oral/nasal ulcers, or alopecia). From a patient perspective is this really low disease activity?

Secondly, although both diseases are associated with fluctuations in disease activity, the cycling rate of flares in SLE is much faster than RA. For instance, among patients with active disease, 1 recent study in lupus found the mean flare rate is 3 flares per year,²⁰ while another in RA found that just over one-third (38%) of patients flared each year.²¹ This rapid rate of flare in SLE makes “landmark-based” outcome measures, those that assess disease activity at discrete moments in time, less accurate. For instance, if the outcome measure is applied at follow-up, in-person visits only, it is possible if not likely that, given this rapid cyclic of flares in SLE, that a flare may be missed entirely. Yet all of the current SLE disease activity indices are landmark-based. Efforts have been made to address this issue by adapting the instruments used such as establishing as a primary outcome measure a threshold percentage of visits during which a patient is in LLDAS.²² The ideal approach would be to apply determine disease activity levels continuously and to calculate the area-under the curve (AUC).A more realistic possibility may be to extend the assessment of disease activity beyond in-person visits by determining the patient acceptable symptom state (PASS) across SLE domains and generating remote, application-based assessments that capture the patient experience of disease activity. Thus, the SLE outcome measure of the future might combine frequent remotely captured assessment of the disease activity from the patient perspective including comparison to predetermined threshold levels of the PASS for each SLE domain with demonstration of the absence of organ threat, including to those organ systems like renal disease that do not bear attendant symptoms, via home urine proteomics testing, to address these challenges.

And finally, for those that view this discussion of future state outcome measures as too esoteric and far-fetched, we turn to a goal has been recognized as important, long held as a theoretical priority, but remains in the bucket of “failing to do what we know.” We must sadly concede that the routine assessment of medication nonadherence among lupus patients remains an element of future, not present, standard of care. Recent examples highlight how critically important this single pivot in the focus of our care is to the 2 of the direst and dreaded outcomes in SLE, rapidly progressive end-stage renal disease and death. In the first case, the only risk factors that identified patient with SLE who progressed to end-stage renal disease (ESRD) in less than 3 years in a large retrospective study of those patients with SLE were patient nonadherence and details of their specific histopathology.²³

In an even more dramatic example of the hazards on nonadherence, antimalarial medication nonadherence among patients with newly diagnosed SLE led to a 71% greater risk of death.²⁴ Yet prior studies²⁵ have indicated that HCQ is taken as prescribed by less than 50% of SLE patients. Furthermore, a recent systematic review²⁶ demonstrated that monitoring HCQ adherence with a simple blood test was clinically useful. Specifically, when HCQ levels were low there was a 3-fold increased likelihood of nonadherence. Evidence of a dose response curve between HCQ blood level and disease activity score was noted with a 58% lower risk of active disease among those with HCQ levels >750ng/ml. Yet the routine monitoring of HCQ levels remains a rarity in clinical practice and specific approaches that have been demonstrated to improve adherence to treatment including motivational interviewing²⁷ are rarely employed. In the end, it is perhaps not surprising that effective medications are more effective when taken effectively. As we progress deeper into the future era of value-based medicine, it is somewhat surprising that we have not attended to the fact that 1 of the ways that a provider can “cheat” the value game is to become expert in encouraging adherence to treatment plans across the rheumatic diseases and by this approach alone, achieve the greater compensation by delivering better patient outcomes.

In the end, delivering on the promise of innovations in lupus care is contingent upon all of us. The practice of real-world Rheumatology must be provided the tools with which to optimize the delivery of meta-omic diagnostic and therapeutic revelations and targeted immunomodulatory pharmacologic advances by seeking to optimize the delivery of care. To do so, translational researchers need to provide adequate outcome measures that capture the complexity of the phenotypic heterogeneity in patient relevant terms at far more frequent intervals. Clinicians must learn to better address non-disease specific limitations to outcome including socioeconomic and racial disparities, behavioral health and other comorbidities and, notably, nonadherence. Health plans and manufacturers have an opportunity to play a role in crafting better care for SLE. By partnering in innovative outcome-based pricing arrangements that link the cost of new medications to the value that they deliver at the individual patient level, cost may be controlled even as new medications with greater effectiveness are introduced. The same overarching approach might be applied to RA. In that disease however, the institution of an effective value driven system must also contemplate how to unwind the established labyrinth of perverse incentives that has resulted in spiraling costs that have grown beyond the control of any single stakeholder, resulted in rising insurance premiums, and to threaten to bankrupt the entire healthcare system.

On that ominous note, time for a reminder: the topic of the next installment will be “A Vision of the Future of RA Care.”

References:

1.Fava A, Buyon J, Mohan C, et al.Integrated urine proteomics and renal single-cell genomics identify an IFN-γ response gradient in lupus nephritis.JCI Insight, 2020; 5(12):e138345. https://doi.org/10.1172/jci.insight.138345.

2.Hoover P, Der E, Berthier CC, et al.Accelerating medicines partnership: organizational structure and preliminary data from Phase 1 studies of lupus nephritis.Arthritis Care Res, 2020; 72(2): DOI 10.1002/acr.24066

3.Humby F, Durez P, Buch MH, et al.Rituximab versus tocilizumab in anti-TNF inadequate responder patients with rheumatoid arthritis (R4RA):16-week outcomes of a stratified, biopsy-driven, multicentre, open-label, phase 4 randomised clinical trial.Lancet, 2021; 397:305-17.

4. Fava A, Buyon J, Mohan C, et al.Urine proteomics and single cell transcriptomics identify IL-16 as a biomarker for lupus nephritis.Abstract 0936, Paper presented at: American College of Rheumatology Convergence 2020.

5. Mejia-Vilet J, Zhang XL, et al.Urinary soluble CD163: a novel noninvasive biomarker of activity for lupus nephritis.JASN, 2020; 31(6); DOI: https://doi.org/10.1681/ASN.2019.2019121285.

6. Aragon CC, Tafur R-A, Suarez-Avellaneda A, et al. Urinary biomarkers in lupus nephritis.J Translational Autoimmunity, 2020.https://doi.org/10.1016./j.auto.2020.100042.

7. Furie R, Rovin BH, Houssiau F, et al.Two-year, randomized, controlled trial of belimumab in lupus nephritis.N Eng J Med, 2020. 383, DOI: 10.1056/NEJMoa2001180.

8. Arriens C, Polyakova S, Adzerikho I, et al.AURORA phase 3 study demonstrates voclosporin statistical superiority over standard of care in lupus nephritis.[abstract taken from Ann Rheum Dis, 2020;http://dx.doi.org/10.1136/annrheumdis-2020-eular.5010].

9. Furie R, Aroca G, Alvarez A, et al.Two-year results from a randomized, controlled study of obinutuzumab for proliferative lupus nephritis.Abstract 0988, Paper presented at: American College of Rheumatology Convergence 2020.

10.A Study to Evaluate the efficacy and safety of obinutuzumab in patients with ISN/RPS 2003 class III or IV lupus nephritis (REGENCY).ClinicalTrials.gov identifier: NCT04221477. https://www.clinicaltrials.gov/ct2/show/NCT04221477.Updated February 18, 2021. Accessed February 26, 2021.

11. Furie RA, Morand EF, Bruce IN, et al.Type I interferon inhibitor anifrolumab in active systemic lupus erythematosus (TULIP-1): a randomised, controlled, phase 3 trial.Lancet Rheumatol, 1(4): DOI: https://doi.org/10.1016/S2665-9913(19)30076-1.

12. Morand EF, Furie R, Tanaka Y, et al.Trial of anifrolumab in active systemic lupus erythematosus. N Engl J Med, 2020; 382, DOI: 10.1056/NEJMoa1912196

13.Safety and efficacy of two doses of anifrolumab compared to placebo in adult subjects with active proliferative lupus nephritis (TULIP-LN1).ClinicalTrials.gov identifier: NCT02547922. https://www.clinicaltrials.gov/ct2/show/NCT02547922.Updated January 27, 2021. Accessed February 26, 2021.

14.Chaichian Y & Strand V.Interferon-directed therapies for the treatment of systemic lupus erythematosus: a critical update.Clin Rheumatol, 2020; https://doi.org/10.1007/s10067-020-05526-1.

15.Lever E, Alves MR, & Isenberg D.Towards precision medicine in systemic lupus erythematosus.Pharmacogenomic Personalized Med, 2020; http://doi.org/10.2147/PGPM.S205079.

16.Rios-Garces R, Espinosa G, van Vollenhoven R & Cervera R.Treat-to-target in systemic lupus erythematosus: where are we?Eur J Int Med, 2020; https://doi.org/10.1016/j.ejim.2020.01.018.

17.Navarro-Compan V, Gherghe AM, Smolen J, et al.Relationship between disease activity indices and their individual components and radiographic progression in RA: a systematic literature review.Rheumatology, 2015; doi:10.1093/rheumatology/keu413

18.Scott IC, Machin A, Mallen CD & Hider SL.The extra-articular impacts of rheumatoid arthritis: moving towards holistic care.BMC Rheumatol, 2018; https://doi.org/10.1186/s41927-018-0039-2.

19.Franklyn K, Lau CS, Navarra SV, et al.Definition and initial validation of a Lupus Low Disease Activity State (LLDAS).Ann Rheum Dis, 2016; http://dx.doi.org/10.1136/annrheumdis-2015-207726.

20.McElhone K, Abbott J, Hurley M, et al.Flares in patients with systemic lupus erythematosus.Rheumatology, 2020; doi:10.1093/rheumatology/keaa777.

21.Rutherford AI, Scott DL, Subesinghe S et al.The Incidence and Predictors of Flare in a Cohort of Rheumatoid Arthritis Patients. [Abstract taken from Rheumatology, 2016; https://doi.org/10.1093/rheumatology/kew132].

22. Sharma C, Raymond W, Eilertsen G & Nossent J.Achieving lupus low disease activity state (LLDAS-50) is associated with both reduced damage accrual and mortality in patients with systemic lupus erythematosus. Arthritis Care Res, 2019; doi: 10.1002/acr.23867

23. Tselios K, Gladman DD, Taheri C, et al.Factors associated with rapid progression to end stage kidney disease in lupus nephritis.J Rheumatol, 2020; doi:10.3899/jrheum.200161

24.Hoque MR, Avina-Zubieta A, De Vera MA, et al.Impact of antimalarial adherence on mortality among patients with newly diagnosed systemic lupus erythematosus: a population‐based cohort study.Arthritis Care Res, 2021; https://doi.org/10.1002/acr.24550.

25.Mehat P, Atiquzzman, Esdaile JM, et al.Medication nonadherence in systemic lupus erythematosus: a systematic review.Arthritis Care Res, 2017;https://doi.org/10.1002/acr.23191

26.Garg S, Unnithan R, Hansen KE, et al.The clinical significance of monitoring hydroxychloroquine levels in patients with systemic lupus erythematosus: a systematic review and meta‐analysis.Arthritis Care Res, 2020; https://doi.org/10.1002/acr.24155.

27. Gossec L, Molto A, Romand X, et al.Recommendations for the assessment and optimization of adherence to disease-modifying drugs in chronic inflammatory rheumatic diseases: a process based on literature reviews and expert consensus.Joint Bone Spine, 2019; https://doi.org/10.1016/j.jbspin.2018.08.006.