Introduction
Cochrane systematic reviews are rigorous in methodology and contribute to our understanding of evidence-based treatments of diseases. Among these diseases is pemphigus, a group of acquired autoimmune vesiculobullous diseases (pemphigus vulgaris [PV], pemphigus foliaceus [PF], and pemphigus paraneoplastic [PNP]), characterized by B-cell–mediated immunoglobulin G antibodies against desmogleins 1 and 3. Cutaneous bullae cause loss of barrier function and pain, dehydration, superimposed infections, and psychological distress. Due to a lack of expert consensus and the poor efficacy of previous therapies, a Cochrane systematic review of randomized controlled trials (RCTs) sought to define the best treatment for pemphigus [1]. Here, we highlight takeaways from the review [1] and discuss advances in therapy. PNP was excluded from the review due to its rarity and because its management depends on the underlying malignancy.
Methods
A total of 11 RCTs were analyzed to assess efficacy and safety among treatments for PV and PF. The primary outcomes were death and disease remission, with secondary outcomes including disease severity indexes, time to disease control, cumulative glucocorticoid dose, serologic markers, and the proportion of patients achieving disease control and relapse. The RCTs used various combinations and doses of steroid-sparing agents with or without corticosteroids. We contrasted therapies, outcomes, and comparison effect size and conducted 4 meta-analyses.
Results
A recent (2021) network meta-analysis [2] found rituximab (Table 1), a CD20 B-cell–depleting therapy, as the most effective therapy for key outcomes like disease relapse, withdrawal from adverse events, remission, and cumulative glucocorticoid dose. The right-most column of Table 1 contrasts therapies relative to rituximab among the 4 key outcomes evaluated. Although the included trials [2] risked bias due to inadequate allocation concealment and lack of participant, personnel, and outcome blinding, the results align with emerging expert consensus and other important clinical trials [3] directly comparing rituximab to other therapies like mycophenolate.
Table 1.
Summary of interventions, outcomes assessed, and effect size from 2009 (left) and 2021 (right).
| 2009 | 2021 | ||||||||||
| Intervention | Effect size: RRa (95% Cl) | Outcomes | Effect size: pooled ORb (95% CI) |
Rituximab vs intervention | |||||||
| Prednisolone (1mg/kg vs 0.5 mg/kg) | Steroid alone | ||||||||||
|
|
1 | Not estimable | Disease control | —c |
|
||||||
|
|
2 | 0.7 (0.43 to 1.14) | Relapse | 0.38 (0.12 to 1.15) |
|
||||||
|
|
3 | Not estimable | Withdrawal due to adverse event | 0.05 (0 to 0.083) |
|
||||||
| Pulsed oral dexamethasone vs placebo | Steroid alone | ||||||||||
|
|
1 | 1.91 (0.68 to 5.33) | Relapse (after discontinuing or stopping) | — |
|
||||||
|
|
2 | 2.45 (0.31 to 19.74) | Withdrawal due to adverse event | — |
|
||||||
| Azathioprine vs glucocorticoid (prednisolone) alone |
|
||||||||||
|
|
1 | 1.04 (0.8 to 1.36) | Remission | 14.45 (4.71 to 43.68) | Steroid alone | ||||||
|
|
2 | –3.91 (–6.71 to –1.12) | Cumulative glucocorticoid dose | –11.10 (–14.08 to –9.57) | Steroid alone | ||||||
|
|
3 | 2 (0.19 to 20.9) | Withdrawal due to adverse event | 0.02 (0 to 0.56) | Azathioprine | ||||||
| Cyclophosphamide vs glucocorticoid (prednisone/prednisolone) alone | Azathioprine | ||||||||||
|
|
1 | 0.96 (0.71 to 1.28) | Remission | 10.10 (2.67 to 38.23) |
|
||||||
|
|
2 | 0 (0) | Disease control | — |
|
||||||
|
|
3 | 0.5 (0.05 to 4.67) | Relapse | 0.60 (0.10 to 3.63) |
|
||||||
|
|
4 | –3.35 (–6.14 to –0.56) | Cumulative glucocorticoid dose | –8.79 (–11.60 to –5.98) |
|
||||||
|
|
5 | 0.33 (0.01 to 7.87) | Withdrawal due to adverse events | — |
|
||||||
| Cyclosporine vs glucocorticoid (prednisone/methylprednisolone) alone | Cyclophosphamide | ||||||||||
|
|
1 | 0 (0) | Remission | 9.59 (2.42 to 37.96) |
|
||||||
|
|
2 | 1.06 (0.86 to 1.32) | Disease control | — |
|
||||||
|
|
3 | 0.92 (0.23 to 3.65) | Relapse | 0.42 (0.08 to 2.28) |
|
||||||
|
|
4 | –0.05 ( –0.18 to 0.081) | Cumulative glucocorticoid dose | –9.36 (–12.16 to –6.55) |
|
||||||
|
|
5 | 0 (0) | Withdrawal due to adverse event | 0.10 (0 to 4.20) |
|
||||||
| Dapsone vs placebo | Cyclophosphamide | ||||||||||
|
|
1 | 1.85 (0.61 to 5.63) | Remission (<7.5 mg prednisone) at 12 months | — |
|
||||||
|
|
2 | 0.37 (0.05 to 2.95) | Withdrawal due to adverse event | — |
|
||||||
| Mycophenolate vs glucocorticoid (prednisolone) alone | Dexamethasone-cyclophosphamide (6 and 12 months) |
||||||||||
|
|
1 | 0.91 (0.67 to 1.24) | Remission | 47.11 (4.99 to 445.07), 6 months |
|
||||||
|
|
2 | –1.83 (–4.94 to 1.28) | Cumulative glucocorticoid dose | — |
|
||||||
|
|
3 | 1.0 (0.07 to 15.26) | Withdrawal due to adverse events | 0.06 (0 to 7.06), 6 months |
|
||||||
| Plasma-exchange vs control | Dexamethasone-cyclophosphamide (6 and 12 months) |
||||||||||
|
|
1 | 7.43 (0.43 to 129.55) | Death | — |
|
||||||
|
|
2 | 1.12 (0.70 to 1.78) | Disease control (study definition involving relative healing time) | — |
|
||||||
|
|
3 | 44.38 (–222.43 to 311.19) | Reduction antibody titer (baseline to end protocol, mean difference) | — |
|
||||||
|
|
4 | 7.2 (0.42 to 124.08) | Withdrawal due to adverse events | — |
|
||||||
| Azathioprine vs cyclophosphamide |
|
||||||||||
|
|
1 | 1.09 (0.82 to 1.44) | Remission | 5.48 (0.71 to 42.02), 12 months | Dexamethasone-cyclophosphamide (6 and 12 months) | ||||||
|
|
2 | 1.8 (0.89 to 3.64) | Disease control (healing of >50% of lesions and/or occurrence of <5 blisters/month) | — | Dexamethasone-cyclophosphamide (6 and 12 months) | ||||||
|
|
3 | 1.0 (0.53 to 1.88) | Relapse | 0.67 (0.04 to 11.13) | Dexamethasone-cyclophosphamide (6 and 12 months) | ||||||
|
|
4 | 1.0 (0.53 to 1.88) | Relapse | 0.063 (0.12 to 3.47) | Mycophenolate | ||||||
|
|
5 | –5.64 (–1.04 to –0.79) | Cumulative glucocorticoid dose | — | Mycophenolate | ||||||
|
|
6 | 3.91 (0.45 to 33.66) | Withdrawal due to adverse events | 0.05 (0 to 1.18) | Mycophenolate | ||||||
| Azathioprine vs mycophenolate | Mycophenolate | ||||||||||
|
|
1 | 1.14 (0.85 to 1.53) | Remission | 10.80 (3.07 to 38.05) |
|
||||||
|
|
2 | 0.72 (0.52 to 0.99) | Disease control | — |
|
||||||
|
|
3 | –2.07 (–3.54 to –0.60) | Cumulative glucocorticoid dose | –11.10 (–13.70 to –8.49) |
|
||||||
|
|
4 | 3.01 (0.48 to 18.97) | Withdrawal due to adverse events | — |
|
||||||
| Cyclophosphamide vs cyclosporine |
|
||||||||||
|
|
1 | 0 (0) | Remission (<10 mg prednisone equivalent) at 5 years | — | Mycophenolate | ||||||
|
|
2 | 0 (0) | Disease control | — | Mycophenolate | ||||||
|
|
3 | 0.4 (0.04 to 3.66) | Relapse | 0.81 (0.05 to 13.72) | Cyclosporine | ||||||
|
|
4 | 0 (0) | Withdrawal due to adverse events | 0.04 (0 to 5.92) | Cyclosporine | ||||||
| Cyclophosphamide vs mycophenolate | Cyclosporine | ||||||||||
|
|
1 | 1.05 (0.76 to 1.44) | Remission | 11.96 (1.92 to 74.49) |
|
||||||
|
|
2 | –1.52 (–2.98 to –0.056) | Cumulative glucocorticoid dose | –11.77 (–14.04 to 9.51) |
|
||||||
|
|
3 | 0.33 (0.01 to 7.87) | Withdrawal due to adverse events | — |
|
||||||
| Topical epidermal growth factor vs placebo | 2.35 (1.62 to 3.41) | Time to control (hazard ratio) | — | Cyclosporine | |||||||
| Traditional Chinese Medicine | 0.75 (–1.12 to 2.62) | Antibody titer | — | Cyclosporine | |||||||
aRR: relative risk.
bOR: odds ratio.
cThe 2021 network review assessed withdrawal due to adverse events, remission, relapse, and cumulative glucocorticoid dose. Other measures were not available.
Induction dosing for rituximab was two 1 g intravenous infusions 2 weeks apart followed by a 6-month prednisone taper of 1 mg/kg/day. Additionally, 2 novel higher-affinity CD20-blocking agents, ofatumumab and veltuzumab, demonstrated efficacy in isolated cases of rituximab-resistant pemphigus. Ofatumumab and veltuzumab are not used for pemphigus outside of clinical trials and for compassionate use. In addition, trials are underway for other immunotherapies targeting the fragment crystallizable region, B-cell–activating factor, and Bruton tyrosine kinase [4]. The meta-analyses revealed that some interventions were superior for certain outcomes: improved disease remission with mycophenolate relative to azathioprine, a steroid-sparing effect with azathioprine and cyclophosphamide, and a decreased time to erosion control with topical epidermal growth factor (Table 2). At the time of the 2009 study, systematic analysis including rituximab and clinical trials including intravenous immunoglobulin were ongoing [5].
Table 2.
Summary of conclusive secondary outcomes (2009).
| Therapeutic | Secondary outcome |
| Mycophenolate mofetil | Improved disease control compared to azathioprine (RRa 0.72, 95% CI 0.52 to 0.99; NNTb 3.7) |
| Azathioprine | Decreased the cumulative glucocorticoid dose (MWDc –3919 mg prednisolone, 95% CI –6712 to –1126) |
| Cyclophosphamide | Deceased the cumulative glucocorticoid dose compared to prednisolone alone (MWD –3355 mg prednisolone, 95% CI –6144 to –566) |
| Topical epidermal growth factor | Decreased time to erosion healing compared to the control intervention (HRd 2.35, 95% CI 1.62-3.41) |
aRR: relative risk.
bNNT: number needed to treat.
cMWD: difference in means.
dHR: hazard ratio.
Discussion
With an increasing understanding of the immune system, B-cell physiology, and the pathogenesis of pemphigus, therapies continue to emerge, making previous therapies obsolete. Here, we placed important Cochrane review findings in the context of recent advancements in the treatment of pemphigus. Further studies are needed to determine therapeutic regimens, safety, and efficacy of novel medical therapies for pemphigus.
Acknowledgments
RPD receives editorial stipends (JMIR Dermatology), royalties (UpToDate), and expense reimbursement from Cochrane. TES receives fellowship funding from Pfizer (grant 25B1519; principal investigator [PI]: Stanca Birlea) and the National Institutes of Health (NIH; grant 2T32AR00741136A1; PI: Dennis Roop). RR receives fellowship funding from the NIH (grant 2T32AR00741131A1; PI: Dennis Roop). The authors would like to thank Linda Martin for reviewing and providing feedback on the manuscript.
Abbreviations
- PF
pemphigus foliaceus
- PNP
pemphigus paraneoplastic
- PV
pemphigus vulgaris
- RCT
randomized controlled trial
Footnotes
Conflicts of Interest: RPD is editor-in-chief of JMIR Dermatology and Cochrane Council cochair. TES serves as an editorial board member-at-large for JMIR Dermatology. RR is an editorial fellow for JMIR Dermatology. DM serves on the advisory boards or is an investigator for ArgenX, Roche, Lilly, Principiabio, Sanofi, and Janssen. DM is also the cocreator of the Pemphigus Disease Area Index and the creator of the Autoimmune Bullous Disease Quality of Life (ABQOL) and Treatment Autoimmune Bullous Disease Quality of Life (TABQOL) questionnaires.
Editorial notice: This article is based on a Cochrane Review previously published in the Cochrane Database of Systematic Reviews 2009, Issue 1, DOI: 10.1002/14651858.CD006263.pub2 (see www.cochranelibrary.com for information). Cochrane Reviews are regularly updated as new evidence emerges and in response to feedback, and Cochrane Database of Systematic Reviews should be consulted for the most recent version of the review.
References
- 1.Martin L, Werth Victoria, Villanueva Elmer, Segall Janet, Murrell Dedee F. Interventions for pemphigus vulgaris and pemphigus foliaceus. Cochrane Database Syst Rev. 2009 Jan 21;(1):CD006263. doi: 10.1002/14651858.CD006263.pub2. [DOI] [PubMed] [Google Scholar]
- 2.Lee M, Yeh Y, Tu Y, Chan TC. Network meta-analysis-based comparison of first-line steroid-sparing adjuvants in the treatment of pemphigus vulgaris and pemphigus foliaceus. J Am Acad Dermatol. 2021 Jul;85(1):176–186. doi: 10.1016/j.jaad.2020.08.028.S0190-9622(20)32417-8 [DOI] [PubMed] [Google Scholar]
- 3.Werth VP, Joly P, Mimouni D, Maverakis E, Caux F, Lehane P, Gearhart L, Kapre A, Pordeli P, Chen DM, PEMPHIX Study Group Rituximab versus mycophenolate mofetil in patients with pemphigus vulgaris. N Engl J Med. 2021 Jun 17;384(24):2295–2305. doi: 10.1056/NEJMoa2028564. [DOI] [PubMed] [Google Scholar]
- 4.Lim YL, Bohelay G, Hanakawa S, Musette P, Janela B. Autoimmune pemphigus: latest advances and emerging therapies. Front Mol Biosci. 2021 Feb 4;8:808536. doi: 10.3389/fmolb.2021.808536. https://europepmc.org/abstract/MED/35187073 .808536 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Atzmony L, Hodak E, Gdalevich M, Rosenbaum O, Mimouni D. Treatment of pemphigus vulgaris and pemphigus foliaceus: a systematic review and meta-analysis. Am J Clin Dermatol. 2014 Dec;15(6):503–15. doi: 10.1007/s40257-014-0101-9. [DOI] [PubMed] [Google Scholar]