Lupus Nephritis, Autoantibody Production and Kidney Outcomes in Males with Childhood-Onset Systemic Lupus Erythematosus

Round 1

Reviewer 1 Report

Very well done study and sheds light to under- studied group. No major suggestions. 

Author Response

Thank you for the kind review.

Reviewer 2 Report

SLE is a very important and serious disease. Especially, the prevalence and incidence of SLE are more in female than male. Therefore, most clinical studies of adult- and childhood-onset lupus are likely biased towards females. This study seeks to characterize the clinical features, natural history and outcomes of male children diagnosed with SLE. Therefore, it is a meaningful paper. This paper is judged to be a well-written paper based on various analysis.

Author Response

Thank you for the kind review.

Reviewer 3 Report

This paper by Wenderfer and colleagues describes  a cohort of 95 male patients with childhood-onset systemic lupus erythematosus (cSLE) and lupus nephritis.

The issue is indeed relevant from a clinical perspective, as cSLE is a rare condition, even more so among male subjects.

I have few criticisms:

- Line 88: “ANA and anti-dsDNA were measured by indirect immunofluorescence and Chritidia luciliae assay respectively”. Both tests are indirect immunofluorescence, the difference is the cell substrate: Hep-2 cells (usually) for ANA and Chritidia luciliae for anti-dsDNA. Please edit accordingly, specifying the substrate for ANA.

- Why anti-dsDNA were not tested with a quantitative test? I believe this provides a study limitation.

- ANA were defined as positive at a dilution of 1:40. This is a very low titer, which should not be regarded as positive. Why did the authors choose such a low cut-off? Please discuss.

- How were aPL tested? Which isotypes were tested? How was aPL positivity defined?

- Please provide positivity rates for anti-beta2GPI antibodies.

- How were ENA detected? Which antibodies were tested?

- CREST is a dated acronym, which is not anymore in use. Please replace it with “systemic sclerosis”.

- What do you mean by “scleroderma”? Localized scleroderma? Systemic sclerosis? Please edit accordingly.

- Please insert further details about lupus non renal involvement, such as the pattern of antibody and clinical manifestations.

-  Please investigate whether low complement and rising anti-dsDNA titers or additional clinical and serological variables might predict renal flares and response to treatment.

- Please detail in the text the female lupus population used as comparison group.

- Was there any patient with multiple flares during follow-up?

- Line 151: Please specify the comparison group: “Males with cSLE had slightly higher rates of…”.

- How many male patients had lupus nephritis at onset? How many females?

- Line 179: “The rate of LN in white males was significant”. I do not get the meaning of this sentence, further details should be added to make it clear to the reader.

- Line 227: Please provide details about the latest follow-up, when SDI was calculated.

- The presence of lupus anticoagulant was higher among patients withot lupus nephritis, but it seems that this was not significant, could you please confirm? Please discuss why aCL and LA were higher in the non lupus nephritis group, as this is not an expected finding.

- It is not clear to me the sentence at line 300: “The rates in males with cSLE were significantly higher”. Males recruited in this study? Compared to which female population?  

- Please specify that gender was self-reported in the Methods.

- Several acronyms are inserted in the text without any details (e.g., DPGN, RR, LN, EKSD, CMS; URSRDS). Please specify the acronyms the first time they are cited in the text.  

Author Response

We appreciate the careful review and the thoughtful recommendations for clarification and improvement of the manuscript. Modifications in the revised version of the manuscript are summarized below, and the line numbers were revised to represent location in the tracked version of the revised manuscript.

Reviewer 3: This paper by Wenderfer and colleagues describes a cohort of 95 male patients with childhood-onset systemic lupus erythematosus (cSLE) and lupus nephritis. The issue is indeed relevant from a clinical perspective, as cSLE is a rare condition, even more so among male subjects.

1) Line 90: “ANA and anti-dsDNA were measured by indirect immunofluorescence and Chrithidia luciliae assay respectively”. Both tests are indirect immunofluorescence, the difference is the cell substrate: Hep-2 cells (usually) for ANA and Chrithidia luciliae for anti-dsDNA. Please edit accordingly, specifying the substrate for ANA.

Thank you for catching this typo. We have revised to say “ANA and anti-dsDNA were measured by indirect immunofluorescence using Hep-2 cells and Chrithidia luciliae assay respectively.”

2) Why anti-dsDNA were not tested with a quantitative test? I believe this provides a study limitation.

Over the years, different laboratories were used and each had non-standardized detection methods for dsDNA reactivity. Therefore, we could not treat dsDNA titer as a quantitative characteristic. This is a limitation of our study and we have included this limitation in the discussion.

3) ANA were defined as positive at a dilution of 1:40. This is a very low titer, which should not be regarded as positive. Why did the authors choose such a low cut-off? Please discuss.

We agree that 1:40 is a low titer, but we are confident that all study participants enrolled had SLE as they met criteria for at least 3 additional ACR criteria for classification of SLE. Most participants had ANA titers >1:1280.

4) How were aPL tested? Which isotypes were tested? How was aPL positivity defined?

5) Please provide positivity rates for anti-beta2GPI antibodies.

Thank you for bringing up this point. Unfortunately, we do not have complete data collected for aPL testing. We have anti-cardiolipin IgG and IgM as well as lupus anticoagulant (DRVVT and Sta-clot). We do not have the data abstracted for anti-beta2 glycoprotein I IgG and IgM. We have clarified this in the methods (lines 102-105), and we have included the lack of anti-beta2GPI antibody data as a limitation in our discussion (lines 328-330). Due to the dramatic difference in aCL reactivity between males with and without nephritis, one of our future directions will be to abstract data on anti-beta2 glycoprotein I IgG and IgM, as well as other antiphospholipid antibodies.

6) How were ENA detected? Which antibodies were tested?

ANA profiles were routinely performed by Missouri labs (https://muhealth.testcatalog.org/show/ANA-Comp). The ANA profiles included anti-Ro/SSA, anti-La/SSB, anti-Sm, anti-RNP, anti-Scl70, and Jo-1. Methods were updated to include this information (lines 94-96).

7) CREST is a dated acronym, which is not anymore in use. Please replace it with “systemic sclerosis”.

We have replaced CREST with systemic sclerosis in line 231 of the text.

8) What do you mean by “scleroderma”? Localized scleroderma? Systemic sclerosis? Please edit accordingly.

The patients diagnosed with scleroderma each had localized scleroderma. We have clarified this in line 231 of the text.

9) Please insert further details about lupus non-renal involvement, such as the pattern of antibody and clinical manifestations.

We included this data in figures 1 and 2 as well as a paragraph in the text, lines 213-220.

10)  Please investigate whether low complement and rising anti-dsDNA titers or additional clinical and serological variables might predict renal flares and response to treatment.

Unfortunately, we did not abstract any clinical data at the time of LN response, nor did we abstract longitudinal data on dsDNA Ab titers, ESR, CRP, IgG levels, or complement C3 or C4 levels between renal response and renal relapse. Therefore, we could not assess whether changes in these parameters predicted future flares in males with cSLE. This is a good idea, and once we can enroll additional males with cSLE into this cohort, we will perform these analyses as a future direction.

11) Please detail in the text the female lupus population used as comparison group.

The primary aim was to describe the male cSLE cohort. Unfortunately, we could only collect a limited set of data on the females with cSLE. But we have not been able to collect as complete a dataset on renal outcomes for our entire female cSLE cohort. We hope to obtain funding to complete the data abstraction on all female sCLS patients for future studies.

12) Was there any patient with multiple flares during follow-up?

Thank you for this suggestion. Actually, 7 of the 8 males with LN flares had multiple. We have added this to the text “Four males developed 2 kidney flares and three more developed 3 flares, for an annualized flare rate of 0.14 flares/person-year in males with cSLE and LN” See lines 233-235. Unfortunately, we do not have clarity of data to report if flares were proteinuric or nephritic.

13) Line 151: Please specify the comparison group: “Males with cSLE had slightly higher rates of…”.

Thank you for pointing out the need for clarification. This sentence relates to the prior sentence comparing males with cSLE and LN to males with non-renal cSLE. We have changed the sentence to read “Males with cSLE and LN had slightly higher rates of…”

14) How many male patients had lupus nephritis at onset? How many females?

Ninety percent of both males and females with cSLE presented with LN, and the remainder developed LN within 2 years of diagnosis. This was included in the text on line 189.

15) Line 179: “The rate of LN in white males was significant”. I do not get the meaning of this sentence, further details should be added to make it clear to the reader.

Thank you for pointing out the need for clarification. The rates of LN class II and III in males were higher and the rates of LN class IV were lower than in females with cSLE. We did not have significant power to assess differences in LN class between both ethnicities and genders. This was corrected.

16) Line 249: Please provide details about the latest follow-up, when SDI was calculated.

The median follow-up for the cohort was 3.4 years, IQR 1.5-6 years, which did not significantly differ between males and females with cSLE, or between patients with or without nephritis. These data are now included in line 245 of the results.

17) The presence of lupus anticoagulant was higher among patients without lupus nephritis, but it seems that this was not significant, could you please confirm? Please discuss why aCL and LA were higher in the non-lupus nephritis group, as this is not an expected finding.

We can confirm that the p-value was 0.15 (non-significant), likely due to the small sample size. We have added the following to the discussion: “The findings that the rates of aCL and lupus anticoagulant were higher in males with cSLE without LN were surprising. No difference has been reported in aSLE populations with and without LN.” Two references were added that assessed aCL positivity in relationship with lupus nephritis in adults with SLE. We prefer not to speculate on why this relationship was seen in cSLE males, but as for Q#4 and Q#5 we plan to collect data on other aPL over time in our cohort as a future direction.

18) It is not clear to me the sentence at line 300: “The rates in males with cSLE were significantly higher”. Males recruited in this study? Compared to which female population?  

Since we did not have these data abstracted on our females with cSLE, we have revised this sentence.

19) Please specify that gender was self-reported in the Methods.

We have revised the sentence to state that “gender (M/F) was self-reported.” Unfortunately, our medical record system restricts gender to either male or female, and additional options were not collected. The point of this statement in the methods was that we collected gender and not biological sex.

20) Several acronyms are inserted in the text without any details (e.g., DPGN, RR, LN, EKSD, CMS; URSRDS). Please specify the acronyms the first time they are cited in the text.

All acronyms have been spelled out in the text (lines 56, 59, 61, 68 and 69) and in table footnotes. Thank you for identifying these omissions.

Round 2

Reviewer 3 Report

Please check typos in the inserted text:

Lupus anticoagulant instead of LA

detection methods FOR dsDNA reactivity