We read with interest the recent manuscript by Neupane et al. (2023) (1) in Lupus Science and Medicine, which compared the efficacy of anifrolumab and belimumab in adults with moderate-to-severe systemic lupus erythematosus (SLE). The study found that anifrolumab and belimumab are comparable in terms of SRI-4 response at 52 weeks using an indirect treatment comparison (ITC) analysis. The authors noted this was different from our ITC (Bruce et al. 2022) (2) findings where patients on anifrolumab were twice as likely to achieve an improvement in SRI-4 as belimumab.
The difference in the results between Neupane et al. and our study might not be an actual difference but explained by differences in data utilised in the ITC analyses.
First, different SRI-4 results were used. Neupane et al. utilised the SRI-4 results derived from pre-specified restricted medication rules for TULIP-1 whereas our study used results derived from amended medication rules (3). There was an error in the pre-specified restricted medication rules which were subsequently amended to ensure clinical appropriateness (3). As highlighted in our Letter in reply (6), the SRI-4 results derived from the amended rules were agreed to by the FDA and the EMA, and reported in the US Prescribing Information (5) and Summary of Product Characteristics (7). Regulators were aligned with the prespecified outcome definition used in TULIP-2 as well as the BLISS trials, thus they were the appropriate rules to use...
We read with interest the recent manuscript by Neupane et al. (2023) (1) in Lupus Science and Medicine, which compared the efficacy of anifrolumab and belimumab in adults with moderate-to-severe systemic lupus erythematosus (SLE). The study found that anifrolumab and belimumab are comparable in terms of SRI-4 response at 52 weeks using an indirect treatment comparison (ITC) analysis. The authors noted this was different from our ITC (Bruce et al. 2022) (2) findings where patients on anifrolumab were twice as likely to achieve an improvement in SRI-4 as belimumab.
The difference in the results between Neupane et al. and our study might not be an actual difference but explained by differences in data utilised in the ITC analyses.
First, different SRI-4 results were used. Neupane et al. utilised the SRI-4 results derived from pre-specified restricted medication rules for TULIP-1 whereas our study used results derived from amended medication rules (3). There was an error in the pre-specified restricted medication rules which were subsequently amended to ensure clinical appropriateness (3). As highlighted in our Letter in reply (6), the SRI-4 results derived from the amended rules were agreed to by the FDA and the EMA, and reported in the US Prescribing Information (5) and Summary of Product Characteristics (7). Regulators were aligned with the prespecified outcome definition used in TULIP-2 as well as the BLISS trials, thus they were the appropriate rules to use for the analyses. The large difference in SRI-4 results between the two definitions – the odds ratio changed from favouring placebo to favouring anifrolumab (0.84 to 1.16) – would have impacted the Neupane et al results had the amended rules been used.
Second, Neupane et al. adjusted the baseline BILAG status from the BLISS trials to match the BILAG status at screening from the TULIP trials. In the TULIP trials, there was a difference in the BILAG status at screening (100% ≥ 1A or 2B) and at baseline (94.4% ≥ 1A or 2B), as expected of a disease with activity that fluctuates over time. Our study used baseline BILAG status from both the TULIP and BLISS trials. It was important to account for this to ensure a fair comparison across the trials.
We believe that these differences in data utilised contributed to the differences in results between the two publications. Had Neupane et al. used the TULIP-1 SRI-4 results derived from the amended medication rules and baseline BILAG status for both sets of trials, the results obtained would have been more consistent with our study.
Finally, Neupane et al. inaccurately claims that most or all differences in results can be explained by our study having insufficient data to undertake their approach. Neupane et al. highlighted the reduction of ESS (Effective Sample Size) to 71 in the Bruce et al. MAIC. However, it should be noted the MAIC in our study was only used as a sensitivity analysis. In the primary analysis with an STC, the sample size was not impacted (710), and results were consistent across both analyses. This showed that these results are robust. Also, the MAIC ESS reduction (90%) was within the range identified by a NICE review (4) where the reduction in ESS ranged from 57% to 98%.
In conclusion, we believe that our study is well-conducted and the results are robust. As we enter a new era of having more choice of therapies for our patients living with SLE, comparative analyses such as this will provide an additional layer of support to clinicians, patients and payers in their decision-making.
Sincerely,
Ian N Bruce MD FRCP
Professor of Rheumatology
Centre for Musculoskeletal Research,
Division of Musculoskeletal and Dermatological Sciences,
Faculty of Biology, Medicine and Health
University of Manchester
Manchester, UK
References:
1. Neupane, S., Sattar, A., Isenberg, D. A., et al. (2023). Belimumab versus anifrolumab in adults with systemic lupus erythematosus: an indirect comparison of clinical response at 52 weeks. Lupus Sci Med., 10(1), e000841.
2. Bruce IN, Golam S, Steenkamp J et al. Indirect treatment comparison of anifrolumab efficacy versus belimumab in adults with systemic lupus erythematosus. J. Comp. Eff. Res. doi:10.2217/cer-2022-0040 (2022).
3. Furie, R. A., Morand, E. F., Bruce, I. N., et al. (2019). Type I interferon inhibitor anifrolumab in active systemic lupus erythematosus (TULIP-1): a randomised, controlled, phase 3 trial. Lancet Rheumatol 2019; 1: e208–19.
4. Phillippo D, Ades T, Dias S et al. NICE DSU technical support document 18: methods for population-adjusted indirect comparisons in submissions to NICE. 2016.
5. SAPHNELO™ Prescribing Information. (2022). https://www.azpicentral.com/pi.html?product=saphnelo
6. Bruce IN, Golam S, Steenkamp K et al. Letter in reply: indirect treatment comparison of anifrolumab efficacy versus belimumab in adults with systemic lupus erythematosus
7. Summary of Product Characteristics – Saphnelo. https://www.ema.europa.eu/documents/product-information/saphnelo-epar-pr...
Ufuk Ilgen
Rheumatology, Ankara University, Ankara, Turkey
On behalf of all co-authors: Ufuk Ilgen, Mucteba Enes Yayla, Askin Ates, Ilyas Ercan Okatan, Emine Uslu Yurteri, Murat Torgutalp, Ayse Bahar Dincer Kelesoglu, Tahsin Murat Turgay, and Gulay Kinikli
Dear Editor,
We read the article by Bao S et al1 with interest. The authors cited our previous work2 by stating: “Prior research of aPL analysis in patients with JSLE did not include the other autoantibodies, and thus the results may not be convincing enough because the impact of the other autoantibodies was neglected in the disease course.” We have some concerns about the interpretation of our results:
1. Our study did not include patients with juvenile-onset SLE. As clearly stated in Methods, all were adult patients. Disease duration did not either date back to childhood (Table 1).
2. Potential impact of other autoantibodies on disease course was not neglected. As stated in Methods, all patients had a panel of 16 autoantibodies (against Sm, nucleosome, histone, PCNA, PO, SS-A 60 kDa, SS-A 52 kDa, SS-B, CENP-B, Scl-70, U1-RNP, Jo-1, PM/Scl, Mi-2, Ku, and dsDNA). However, none were found to be associated with antiphospholipid antibody status.
3. The aim of our study was not to determine a relationship between a general autoantibody profile and SLE manifestations at all. Patients with serial antiphospholipid antibody measurements were included in the study.
Some additi...
Ufuk Ilgen
Rheumatology, Ankara University, Ankara, Turkey
On behalf of all co-authors: Ufuk Ilgen, Mucteba Enes Yayla, Askin Ates, Ilyas Ercan Okatan, Emine Uslu Yurteri, Murat Torgutalp, Ayse Bahar Dincer Kelesoglu, Tahsin Murat Turgay, and Gulay Kinikli
Dear Editor,
We read the article by Bao S et al1 with interest. The authors cited our previous work2 by stating: “Prior research of aPL analysis in patients with JSLE did not include the other autoantibodies, and thus the results may not be convincing enough because the impact of the other autoantibodies was neglected in the disease course.” We have some concerns about the interpretation of our results:
1. Our study did not include patients with juvenile-onset SLE. As clearly stated in Methods, all were adult patients. Disease duration did not either date back to childhood (Table 1).
2. Potential impact of other autoantibodies on disease course was not neglected. As stated in Methods, all patients had a panel of 16 autoantibodies (against Sm, nucleosome, histone, PCNA, PO, SS-A 60 kDa, SS-A 52 kDa, SS-B, CENP-B, Scl-70, U1-RNP, Jo-1, PM/Scl, Mi-2, Ku, and dsDNA). However, none were found to be associated with antiphospholipid antibody status.
3. The aim of our study was not to determine a relationship between a general autoantibody profile and SLE manifestations at all. Patients with serial antiphospholipid antibody measurements were included in the study.
Some additional concerns are:
a. If patients with juvenile-onset SLE in the study by Bao S et al are divided into two groups based solely on the anti-U1-RNP status, the generated groups will exactly be the same as the cluster analysis results provided in Table 2. If a cluster analysis by just a single variable, the anti-U1-RNP status, is run, one will also have exactly the same result. Similar to antiphospholipid antibodies, the inclusion of autoantibodies other than the anti-U1-RNP to clustering would not have any impact on the results. So, the results provided in Table 3 may be interpreted as “clinical features associated with anti-U1-RNP antibodies”.
b. Some autoantibodies such as anti-dsDNA, anti-Sm, and anti-RNP have been known to have strong clinical associations, whereas some have little impact on disease phenotype and activity. Moreover, each may have somewhat differential impact on disease manifestation domains. As an example, positivity (including positive seroconversion) of anti-dsDNA has been known to be associated with nephritis which strongly contributes to disease activity, whereas anti-La positivity (including positive seroconversion) is associated with sicca which is not considered as a component of SLE disease activity in widely used disease activity indices such as SLEDAI. Putting all seroconversions in a single pool may cause losing fine clinical associations of particular autoantibodies. High rate of lupus nephritis in seroconverters in this study is not necessarily related to seroconversion itself, rather this may be related to high prevalence of baseline anti-Sm/RNP in the seroconverter group (Table 1). This point requires further analysis.
In conclusion, clustering based solely on autoantibody profile may not have a clinical translation other than established clinical associations of particular autoantibodies. Clinical associations of seroconversion should be interpreted for each particular autoantibody.
References
1. Bao S, Huang H, Jin Y, et al. Autoantibody-based subgroups and longitudinal seroconversion in juvenile-onset systemic lupus erythematosus. Lupus Science & Medicine 2023;10:e000834. doi:10.1136/ lupus-2022-000834
2. Ilgen U, Yayla ME, Ateş A, et al. Antiphospholipid antibodies and non-thrombotic manifestations of systemic lupus erythematosus. Lupus 2018;27:665-9. doi: 10.1177/0961203317734924
We read with interest the recent manuscript by Neupane et al. (2023) (1) in Lupus Science and Medicine, which compared the efficacy of anifrolumab and belimumab in adults with moderate-to-severe systemic lupus erythematosus (SLE). The study found that anifrolumab and belimumab are comparable in terms of SRI-4 response at 52 weeks using an indirect treatment comparison (ITC) analysis. The authors noted this was different from our ITC (Bruce et al. 2022) (2) findings where patients on anifrolumab were twice as likely to achieve an improvement in SRI-4 as belimumab.
The difference in the results between Neupane et al. and our study might not be an actual difference but explained by differences in data utilised in the ITC analyses.
First, different SRI-4 results were used. Neupane et al. utilised the SRI-4 results derived from pre-specified restricted medication rules for TULIP-1 whereas our study used results derived from amended medication rules (3). There was an error in the pre-specified restricted medication rules which were subsequently amended to ensure clinical appropriateness (3). As highlighted in our Letter in reply (6), the SRI-4 results derived from the amended rules were agreed to by the FDA and the EMA, and reported in the US Prescribing Information (5) and Summary of Product Characteristics (7). Regulators were aligned with the prespecified outcome definition used in TULIP-2 as well as the BLISS trials, thus they were the appropriate rules to use...
Show MoreUfuk Ilgen
Rheumatology, Ankara University, Ankara, Turkey
On behalf of all co-authors: Ufuk Ilgen, Mucteba Enes Yayla, Askin Ates, Ilyas Ercan Okatan, Emine Uslu Yurteri, Murat Torgutalp, Ayse Bahar Dincer Kelesoglu, Tahsin Murat Turgay, and Gulay Kinikli
Dear Editor,
We read the article by Bao S et al1 with interest. The authors cited our previous work2 by stating: “Prior research of aPL analysis in patients with JSLE did not include the other autoantibodies, and thus the results may not be convincing enough because the impact of the other autoantibodies was neglected in the disease course.” We have some concerns about the interpretation of our results:
1. Our study did not include patients with juvenile-onset SLE. As clearly stated in Methods, all were adult patients. Disease duration did not either date back to childhood (Table 1).
2. Potential impact of other autoantibodies on disease course was not neglected. As stated in Methods, all patients had a panel of 16 autoantibodies (against Sm, nucleosome, histone, PCNA, PO, SS-A 60 kDa, SS-A 52 kDa, SS-B, CENP-B, Scl-70, U1-RNP, Jo-1, PM/Scl, Mi-2, Ku, and dsDNA). However, none were found to be associated with antiphospholipid antibody status.
3. The aim of our study was not to determine a relationship between a general autoantibody profile and SLE manifestations at all. Patients with serial antiphospholipid antibody measurements were included in the study.
Show MoreSome additi...