Assessment of ovarian reserve in reproductive age Indian women with lupus and determination of correlates of diminished reserve: results from a tertiary care centre

Introduction

Systemic lupus erythematosus (SLE) is the prototype systemic autoimmune rheumatic disease which predominantly affects women of reproductive age group; its reported global population prevalence is 43.7 (15.9–108.9) per 100,000 persons with a female: male ratio of about 9:1 (1). While the reported Indian prevalence of 3.2 (0–6.86) per 100,000 is lower than the global prevalence, the sketchy data dates back to more than three decades ago and is a potential underrepresentation of the true disease burden (2). Nevertheless, even with these conservative estimates, the burden of lupus in India remains significantly high considering the large population size. Fertility, which is an integral component of the overall quality of life of a woman in childbearing age group, remains a controversial issue in SLE. Fertility is believed to be largely preserved in women with SLE, though a subset experiences subfertility due to multiple interacting factors including exposure to gonadotoxic therapy [particularly cyclophosphamide (CYC) at high cumulative doses], delayed conception at advanced age (often to achieve disease remission), effect of chronic kidney disease, menstrual irregularities, and various psychosocial factors (3–5). Although a higher incidence of premature ovarian failure [now better termed premature ovarian insufficiency (POI)] has been shown in patients with SLE in some studies, most of this increased risk has been linked to the prior use of CYC (3, 6–8). Whether SLE per se has a direct impact on fertility and what is the exact magnitude of this impact, remains unclear (3–5).

Fertility is directly related to a woman's ovarian reserve, a measure of the oocyte or follicular pool remaining in her ovaries. Ovarian reserve thus serves as a good surrogate marker for assessment of fertility. Assessment of ovarian reserve is also important for reproductive planning, especially since many patients delay conception for proper control of disease activity or due to personal choices. It is also needed in the planning and execution of assisted reproductive technologies, which are increasingly being employed in patients with lupus (9). Ovarian reserve can be assessed biochemically [through estimation of levels of anti-mullerian hormone (AMH), follicle stimulating hormone (FSH), estradiol (E2), or inhibin B in blood] or through ultrasonographic assessment of antral follicular count (AFC) and ovarian volume (OV). Among these, AMH and AFC are considered the best markers, since both accurately reflect the oocyte pool, and AMH has the added advantage of being more sensitive and cycle independent unlike FSH and E2 (10).

Small studies on AMH as an ovarian reserve marker in SLE have yielded somewhat conflicting results; most (11–15), but not all (16, 17), have shown reduced serum AMH in patients with lupus compared to healthy controls but with a high degree of heterogeneity as shown in a recent systematic review and meta-analysis (18, 19). Significant variations in ovarian reserve due to race and ethnicity have been previously demonstrated in healthy (fertile or infertile) women, and there is preliminary data to suggest that healthy Indian women undergo accelerated ovarian ageing (20–22). However, this has not been studied till date in lupus and it remains unclear whether there are any ethno-geographic variations in fertility or ovarian reserve in women with SLE. More importantly, factors affecting ovarian reserve in lupus remain unclear. While increasing age and prior exposure to CYC, especially at high doses, have been the most consistently identified associations, there is conflicting or no data on the association with disease activity, severity, disease damage, or autoantibodies (3, 4). Data on other ovarian reserve markers such as AFC and OV is sketchy (12, 15, 23, 24). To the best of our knowledge, there is no data on ovarian reserve in Indian lupus patients, despite a high nationwide disease burden. During the ongoing randomized clinical trial (PROGONAD-CP, registered CTRI/2023/11/060157) assessing the efficacy of leuprolide acetate in preventing cyclophosphamide-induced ovarian toxicity in Indian SLE patients, we observed a large number of pre-trial exclusions due to low baseline AMH level (<1 ng/mL) even in young unmarried females with SLE (25).

Therefore, in the present study, we aimed to comprehensively assess the fertility and ovarian reserve in reproductive age Indian SLE females and identify clinico-immunological correlates of reduced ovarian reserve in our patients.

Materials and methods

This single-centre, observational study was done at the All India Institute of Medical Sciences, New Delhi between June 2023 to June 2025. Consecutive, adult female patients in the reproductive age group (18–45 years) classified with SLE as per the Systemic Lupus International Collaborating Clinics (SLICC) 2012 criteria and/or the 2019 American College of Rheumatology (ACR)/European League of Associations for Rheumatology criteria (26, 27) and being treated on out/in-patient basis under the Department of Medicine or Rheumatology were included after a written, informed consent. Those with other gynaecological disorders which could affect the ovarian reserve such as endometriosis or polycystic ovary syndrome and those with a history of prior ovarian surgeries or use of chemotherapy or radiotherapy were excluded (Supplementary Figure S1). The study was approved by the Institutional Ethics Committee (IEC-645/06102023, RP-27/2023).

Data pertaining to the patients' demographic profile, comorbidities, marital status, disease-related parameters such as organ involvement, disease duration, disease activity at the time of assessment, autoantibody profile, and treatment details (including prior exposure to CYC, and its cumulative dose received) were noted in a pre-designed proforma. Disease activity was assessed using Systemic lupus erythematosus disease activity index 2000 (SLEDAI-2K) (28); active disease defined as SLEDAI-2K≥4. Disease damage was assessed using the SLICC/ACR damage index (29). Data on antinuclear antibodies (ANA), extractable nuclear antigen profile including anti-double stranded deoxyribonucleic acid (anti-dsDNA), anti-Smith (anti-Sm), anti U1-ribonucleoprotein (anti-U1RNP), anti-Ro60, anti-Ro52, and anti-La, antiphospholipid antibodies including anticardiolipin (aCL IgM, IgG), anti-beta-2-glycoprotein-I (anti-b2gpI IgM, IgG) and lupus anticoagulant (LAC), and anti-thyroid peroxidase antibody, where available, were recorded. Details about menstrual and obstetric history, including contraceptive usage (where available), were noted. In case of menstrual irregularities, duration of irregularity was also noted. POI was defined as the loss of ovarian function, defined by sustained amenorrhea for at least 6 months with elevated FSH > 40 mIU/mL, before the age of 40 years, while the term early menopause was used for women losing ovarian function in the age group 40–44 years as per the current recommendations (30).

A holistic assessment of fertility-related parameters including number of pregnancies (both before and after disease onset) and their outcomes, and a detailed assessment of ovarian reserve using biochemical markers including AMH, FSH, and estradiol, and ultrasonographic measures including AFC and OV (for a subset of patients) was done. Testing for ovarian reserve markers was done in all patients early in the follicular phase (day 2 to day 5 of menses) wherever feasible; in patients with prolonged amenorrhea, testing was done immediately after enrolment into the study. Peripheral venous blood was collected in plain vials and after clot formation, serum was separated by centrifugation. All hormones were measured using chemiluminescence-based immunoassays on autoanalyzers (AMH—Access 2, Beckman Coulter; FSH, and estradiol—Architect, Abbott Healthcare). For a subset of patients in both groups, sonographic assessment of AFC and OV was done using transvaginal or transabdominal sonography by trained personnel (JB). Antral follicles measuring between 2 mm to 9 mm were visualized in all three planes in each ovary and counted. The total count of both ovaries was considered AFC_total, while the higher value on either side was taken as AFC_max. Ovarian volume on either side was calculated in cubic centimetre as per the ellipsoid formula (0.526 × length × height × width); the average of the two (corresponding to the mean ovarian volume) was calculated for the purpose of analysis.

Statistical analysis

Data were entered in an Excel spreadsheet and analyzed using Statistical Package for Social Sciences [SPSS Inc., Chicago, IL, version 26]. Categorical variables were summarized as n (%) and analyzed using chi-square test or Fisher's exact test. Normality assessment of continuous data was done using Kolmogorov–Smirnov and Shapiro Wilk test; parametric data was reported as mean (SD) and analyzed through student's t test, while non-parametric data was reported as median (IQR) and analyzed through Mann–Whitney U test. Proportion of patients with POI and early menopause were estimated. The characteristics of patients with low AMH (defined as AMH < 1 ng/mL) were compared with those having normal (≥1 ng/mL) serum AMH to identify patient-, disease-, and treatment-related factors associated with reduced ovarian reserve in Indian women with lupus. Sensitivity analysis was performed using a higher AMH cut-off of 1.2 ng/mL (as per the POSEIDON criteria) to define low vs. normal AMH groups (31). Logistic regression was performed to identify determinants of low ovarian reserve; autoantibodies and disease activity were included as dichotomous variables for regression analysis. Nagelkerke R square and Hosmer and Lemeshow Test were used for assessment of goodness-of-fit of the regression model. Recently published age-specific normative data of AMH and AFC from 1,338 healthy, fertile women from our centre was used for comparison (22). Since this normative data included women in the age group of 20–40 years, the nearest available centile values were used for adjudication in women aged 18–19 years and 41–44 years. A two-tailed p-value <0.05 was considered to define statistical significance. A sample size of convenience was chosen and consecutive sampling technique was followed.

Results

Baseline demographic data and clinico-laboratory profile of lupus

A total of 168 women with mean (SD) age 27 (6.4) years were enrolled into the study. 93 (55%) of the included patients were married; of these, two were currently separated or divorced. 45 (27%) patients had one or more comorbidities, most commonly hypothyroidism (35, 21%), followed by hypertension (13, 8%). One (0.6%) each diabetes mellitus, acquired immunodeficiency syndrome, and adrenal incidentaloma. None reported smoking or alcohol use.

149 (89%) had pure lupus, while 19 (11%) had overlap with inflammatory myositis (n = 12), systemic sclerosis (n = 3), both (n = 2), generalized morphea (n = 1) or primary Sjogren's syndrome (n = 1). The median (range) duration of disease at the time of enrollment was 24 (1–172) months. 123 (73%) patients had active disease (SLEDAI ≥4) while 40 (24%) had SLEDAI <2. Renal involvement was seen in 102/167 (61%), of which 88 (53%) had biopsy proven lupus nephritis and 7 had gone into chronic kidney disease [four (57%) stage V, one (14%) each stage II, III and IV]. Anti-dsDNA antibodies were seen in 112/164 (68%) patients, while anti Ro-60 antibodies were positive in 75/159 (47%). 43/141 (30.5%) had aPL positivity (any one or more of the three criteria aPLs), while LAC positivity was seen in 20/117 (17%). C3 was low in 129/163 (79%) while C4 was low in 98/164 (60%) anytime during the disease course. Details about organ involvement and autoantibody profile of included patients are given in Table 1. Twenty (12%) patients had received prior CYC, mean (SD) cumulative dose being 3,636 (1,616) mg.

Table 1. Baseline demographic data, clinico-laboratory profile, treatment details including cyclophosphamide exposure, and menstrual and obstetric details of the study population.

Menstrual and obstetric data

Menstrual cycles were regular in 120/167 (72%) patients; 13/167 (8%) patients had oligomenorrhea, 32/167 (19%) patients had amenorrhea [median (IQR) duration of 4 (4) months], while one each (0.6%) had polymenorrhea or menorrhagia. Three (1.8%) patients had POI (at 21, 22, and 37 years of age), while two (1.2%) fit the definition for early menopause (41 and 44 years, respectively)- only one of these reported prior CYC use (4 g); none had CKD. The mean (SD) age at menarche was 13 (1) years.

Of the 93 married females, 5 had primary infertility (duration 24–132 months), 8 had not planned conception yet, 2 were separated or divorced, and 2 had ongoing pregnancy at the time of writing this report. In the remaining 76 patients, 205 pregnancies were reported; most of these (194, 95%) were before lupus onset. Most pregnancies (143, 70%) resulted in a live birth. Four still births and 40 spontaneous abortions occurred; MTP was performed in 22 (Table 1).

Contraception of any form was used only in 36/91 (40%) married females (Table 1); in the remaining, only 3 (5%) reported having been given contraceptive advice by their treating physician.

Hormonal assessment, antral follicular count and ovarian volume

Serum levels of AMH, FSH, and E2 in the total study population are detailed in Table 2. 82 (49%) patients had low AMH (<1 ng/mL). 87 (52%) patients had low AMH as per the cutoff of <1.2 ng/mL used in the Poseidon criteria. 52 (31%) patients had very low serum AMH (<0.5 ng/mL). Patients in the low AMH group had significantly lower E2 levels; FSH levels were however similar in the two groups (Table 2).

Table 2. Biochemical (AMH, FSH and E2) and sonographic markers of ovarian reserve in the study population and in the two groups [data represented as median (IQR)].

Ovarian USG for AFC and OV was performed for 72 patients, 35 with low AMH and 37 with normal AMH. AFC_total, AFC_max, and mean OV were significantly lower in patients in the low AMH group (Table 2). 29 (40%) patients had AFC_max less than 5; while 39 (55%) had AFC_total less than 10. AFC showed a moderate positive correlation with serum AMH (Figure 1).

Since previous studies and meta-analyses have reported these parameters as mean (SD) despite a non-normal distribution, we have also provided the mean (SD) values in Supplementary Table S1 for ease of comparison.

Comparison with age-specific nomograms of AMH and AFC from healthy fertile women

In 160 women (representing 95% of the study cohort), AMH was lower than the median (50th centile) age-specific AMH seen in healthy fertile women; of these, 50 (30%) had AMH levels lower than the corresponding third centile, while 107 (64%) had AMH less than the corresponding tenth centile of AMH in healthy fertile women.

Similar comparisons for AFC showed that the observed AFC was lower than the median (50th centile) age-specific AFC of healthy fertile women in 54/70 of our patients (representing 77% of the study cohort). Of these, 17 (24%) had AFC less than the 3rd centile, while 33 (47%) had AFC less than the corresponding 10th centile of AFC in healthy fertile women.

Comparison of lupus patients having low AMH with those having normal AMH

Patients of lupus with low vs. normal AMH were compared to identify determinants and correlates of low ovarian reserve. Patients with low AMH were older [29 (7.1) years] compared to those with normal AMH [25.2 (4.9) years, p = 0.001], had more menstrual irregularities (43% vs. 14%, p < 0.001), and a higher proportion of patients with active disease (84% vs. 63%, p = 0.002) (Table 3). Since age is the strongest predictor of ovarian reserve, age-adjusted analysis was performed for these parameters (Table 4).

Table 3. Comparison of patient demographics, clinico-laboratory profile, and CYC exposure among patients in the low AMH and normal AMH groups.

Table 4. Unadjusted and age-adjusted odds ratio of potential predictors of low AMH (<1 ng/mL) identified through regression analysis.

Values in bold represent statistically significant p-values.

There was no difference in clinical profile between the two groups; notably, the proportion of patients with LN or CKD was similar across the two groups. Both groups were similarly exposed to CYC in the past. Among the autoantibodies, Anti Ro was found to be more common in patients with low AMH. Rest of the autoantibodies including dsDNA, Sm, U1RNP were similar in the two groups. Antiphospholipid antibodies were similarly distributed in both groups. Low complements at any time during the disease course were more commonly seen in patients with low AMH compared to those in the normal age group (Table 3).

Regression analysis for predictors of low AMH

A regression analysis was subsequently performed to identify potential predictors of low AMH in Indian lupus patients. Unadjusted and age-adjusted ORs with p-values were first calculated (Table 4). Age, marital status, menstrual irregularities, disease activity, low C3, and anti-Ro60 positivity were found to be associated with low AMH in unadjusted (univariate) analysis. Adjustment for age nullified the association with marital status found on univariate analysis, while adjustment for disease activity nullified the association with low C3 and low C4. After stepwise multivariate regression, higher age [OR: 1.12 (1.05–1.20, p = 0.001], active disease [OR: 3.87 (1.56–9.62), p = 0.004], menstrual irregularities [OR: 4.56 (1.95–10.65), p < 0.001], and anti-Ro60 positivity [OR: 1.89 (0.92–3.88), p = 0.08] were found to be independently associated with reduced AMH in our SLE patients (Table 5). Nagelkerke R² of the model was 0.32; Hosmer and Lemeshow test p-value was 0.34, suggesting a good fit.

Table 5. Final multivariable regression model for prediction of low AMH (<1 ng/mL) in patients with SLE.

A sensitivity analysis for comparison was also performed using an AMH cutoff of <1.2 ng/mL used to define low AMH as per the Poseidon criteria and similar results were obtained (Supplementary Table S2).

Discussion

This single-centre, observational study shows that Indian patients with lupus have a significantly diminished ovarian reserve even in the absence of prior gonadotoxic therapy or other known risk factors. The study also delves into the clinico-immunological characteristics of these patients and identifies novel correlates and predictors of diminished ovarian reserve in female SLE patients through multivariable regression analysis.

Fertility in lupus has not been studied comprehensively and many controversies continue to exist. While one school of thought believes that fertility is largely unaffected primarily due to the disease unless patients have disease complications such as chronic kidney disease or prior therapy with CYC, unequivocal evidence from our study and recent metanalyses which show reduced ovarian reserve in patients with lupus compared to healthy controls and sporadic evidence of autoimmune oophoritis previously demonstrated seem to support the counter-view (18, 19, 32).

Studies on AMH, the most studied ovarian reserve marker in SLE, have showed somewhat conflicting results- while some have reported no difference (4, 16, 17, 33), most have reported lower AMH levels in lupus patients compared to controls (11–15, 23, 24). The variability is evident in the high degree of heterogeneity (I² = 90%) in the pooled estimate of these results in a recent meta-analysis (19). Whether this variability is due to differences in population characteristics or testing methodology, or reflect a potential ethnogeographic variation in ovarian reserve in women with lupus is unclear. Significant variation in ovarian reserve has previously been demonstrated in healthy women of various ethnicities but there is no such systematic study on patients with lupus (20). The ovarian reserve in Indian women has been shown to decline earlier compared to their western counterparts (21). However, there was hitherto no data on ovarian reserve in patients with lupus from the Indian subcontinent. Our study bridges this lacuna in published literature and shows that Indian lupus patients have a significantly reduced ovarian reserve. However, multinational comparative studies will be needed to better study the potential ethno-geographic variations in fertility and ovarian reserve in women with lupus.

Studies on sonographic markers of ovarian reserve are very limited (12, 14, 15, 23, 24). We measured AFC and OV for a subset of patients in both groups and found a good correlation with AMH, adding robustness to our study findings. Our findings of reduced AFC in patients with SLE are in line with the pooled estimate of AFC in SLE derived in a recent metanalysis (19). There is limited data on ovarian volume, another well-established ovarian reserve marker, in patients with lupus (12, 23); our study findings are in concurrence with the limited available data and suggest that ovarian volume is indeed reduced in patients with SLE.

Comparison of our findings with recently published age-specific normative data of AMH and AFC derived from a large cohort of 1,338 healthy women from our centre (22) clearly demonstrates that Indian patients with SLE have significantly lower AMH and AFC values compared to the general population. Since our patients had very little exposure to CYC or hormonal OCPs, and other factors which could affect ovarian reserve (such as endometriosis, genitourinary TB, PCOS, prior radiation or surgery) were excluded, this almost directly incriminates lupus disease itself as the cause of significantly reduced ovarian reserve in these patients.

We explored the clinico-immunological correlates and determinants of diminished ovarian reserve in our patients and found that there was no difference in the organ involvement (including lupus nephritis) between patients with reduced or preserved AMH. Prior CYC exposure was low and similar in both the groups. While there is some published data on ovarian reserve in SLE patients as detailed above, only a few studies have tried looking into the contributory factors. Increasing age has been the most reproducible determinant of reduced ovarian reserve in the studies done so far, consistent with the expected age-related decline in ovarian reserve, and as also seen in our study (19). While prior CYC exposure (especially at higher cumulative doses, and in adults) is believed to be associated with reduced ovarian reserve (18, 34, 35), results have been conflicting with some studies reporting no association with prior CYC exposure (17, 24, 36). Additionally, lupus patients who have not had prior CYC exposure have also been found to have a diminished ovarian reserve compared to age-matched healthy controls, suggesting interplay of other factors currently poorly understood. While some studies have found an association with disease activity, disease damage, duration, peripheral blood lymphocyte subsets, and ESR, these findings have been inconsistent and non-replicative (3, 4). We found a significant association of active disease with reduced serum AMH levels, similar to the findings of Gao, Girbash, Martins, and Alamdarloo et al. (13, 15, 23, 37). Contrarily, no association with disease activity was found in a few other studies (11, 17, 24, 33, 38). Since a cross-sectional assessment of disease activity may not appropriately capture its effect on a parameter like ovarian reserve which depends on the follicular pool, a longitudinal, timed assessment of disease activity and the disease damage accrual over time would be better suited to study the precise association between these parameters. Studying disease activity as a continuous variable vis-a-vis as a dichotomous variable may also have a bearing on the observed association or a lack thereof. We did not find any association with disease duration, similar to other studies which have looked at this association (11, 17).

Most studies on association have focused on clinico-demographic factors and treatment exposures with little emphasis on SLE-related autoantibody profiles. We found positivity for anti-dsDNA, anti-Sm, anti-U1RNP, and aPLs to be similar in patients with reduced and normal AMH; however, we could not find any previously published study which has studied association of ovarian reserve in SLE with these autoantibodies for comparison. As far as aPLs are concerned, their role in infertility and invitro fertilization (IVF) success has long been under debate (39, 40). While a recent study in patients with SLE and APS showed a greater reduction in ovarian reserve (as measured by AMH and AFC) in patients with SLE and secondary APS compared to SLE and primary APS alone, this observation has not been consistently replicated (41).

Interestingly, we found that a higher proportion of patients with reduced AMH had anti-Ro60 antibody positivity. The association persisted even after adjustment for age. No published ovarian reserve study in SLE has directly analyzed or reported correlations between anti-Ro/SSA (Ro60/Ro52) antibodies and ovarian reserve markers. Cohorts of primary Sjögren's syndrome have reported early menopause and there are sporadic case reports of premature ovarian insufficiency, suggesting earlier ovarian aging in these patients (42, 43). Anti-Ro/SSA has also been associated with adverse IVF and pregnancy outcomes (44). The observed association between anti-Ro60 antibodies and reduced ovarian reserve is exciting because of its biologic plausibility. Ro antigens (Ro52/TRIM21 and Ro60) are intracellular RNA-binding proteins that are expressed in the ovarian tissue in various cell types and may get exposed through epitope spreading during follicular apoptosis, which can then allow anti-Ro antibodies to bind and form immune complexes, amplifying local inflammation and accelerated follicular loss. Alternatively, anti-Ro antibodies, which are known to dysregulate IFN pathways, can promote pro-apoptotic signalling and lead to follicular atresia and premature ovarian insufficiency. While our observation is novel and lays insight into the potential mechanisms, it nevertheless needs to be confirmed in future studies with larger sample size from other large centres around the world.

The study suffers from many limitations. First, this was a single-centre study, although being the apex tertiary care referral centre, patients from all over the country are referred to our centre. Second, controls were not included as a part of this study, since age-specific normative data of AMH and AFC derived from 1,338 healthy patients from our centre were available for comparison. However, while the AFC assessment method was identical across the two studies, the normative data for AMH measurement were derived using a second-generation ELISA, whereas the current study utilized the more recent automated chemiluminescent immunoassay. Although automated AMH assays offer advantages such as higher analytical precision, faster turnaround time, reduced inter-assay variability, and lower limits of detection, they may not be directly interchangeable with ELISA-based methods due to assay differences. Nevertheless, published studies have demonstrated strong correlation between automated assays and second-generation ELISA, and pilot work conducted in our laboratory on a small representative sample set showed that inter-assay differences were clinically insignificant and unlikely to influence the results of the present study. Third, antibodies were studied as dichotomous variables since method of their testing was line immunoassay due to which antibody titres were not available. Fourth, longitudinal assessments of disease activity, complements, and AMH levels were not performed. Fifth, the study cohort had a high proportion (73%) of patients having active disease (SLEDAI≥4), which could potentially affect the generalizability of the study findings to patients with milder or inactive disease. Sixth, since only 12% of the study cohort had prior CYC exposure, our study may be underpowered to study an association (or a lack thereof) between CYC exposure and ovarian reserve in Indian lupus patients.

Our study also has some strengths. This is the first study demonstrating low ovarian reserve in Indian lupus patients. Second, a holistic assessment of fertility and ovarian reserve has been done, employing all known biochemical and sonographic parameters. Third, the immunological characteristics of lupus patients with diminished ovarian reserve have been systematically assessed and compared with those having a preserved ovarian reserve for the first time. Fourth, a fairly large sample size was included. Fifth, multivariate regression analysis was performed to identify independent predictors of reduced ovarian reserve in patients with SLE.

However, as is evident from the performance characteristics of the regression model, much still remains unknown regarding the factors affecting ovarian reserve in patients with lupus. Data on anti-ovarian antibodies remains sketchy and conflicting. Whether these patients harbor an underlying genetic predisposition needs to be explored. Lastly, given the observed (negative) association of ovarian reserve with disease activity and the well-recognized possibility that autoimmune POI may be at least partially reversible over time (and with treatment), it remains to be explored whether a similar principle applies to serial measurements of serum AMH and other ovarian reserve biomarkers across different time points and disease activity states during the disease course. Results of our ongoing randomized clinical trial (PROGONAD-CP, registered CTRI/2023/11/060157) hope to shed more light on this subject (25).

In conclusion, we observed a high proportion (49%) of Indian SLE patients had evidence of low ovarian reserve even in the absence of known risk factors, suggesting a potential direct influence of lupus on ovarian function. Higher age, active disease, menstrual irregularities, and anti-Ro60 positivity were found to be independently associated with diminished ovarian reserve in our patients. No association was found with prior CYC use, lupus nephritis, chronic kidney disease or antiphospholipid antibodies. Large, multinational studies including lupus patients from different ethnogeographic settings are needed to validate the novel findings of our study.

Generative AI statement

The author(s) declared that generative AI was not used in the creation of this manuscript.

Any alternative text (alt text) provided alongside figures in this article has been generated by Frontiers with the support of artificial intelligence and reasonable efforts have been made to ensure accuracy, including review by the authors wherever possible. If you identify any issues, please contact us.

Publisher's note

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.

Supplementary material

The Supplementary Material for this article can be found online at: https://www.frontiersin.org/articles/10.3389/flupu.2026.1783078/full#supplementary-material

Supplementary material Supplementary material at the source

Footnotes

Edited by: Klara Rosta, Medical University of Vienna, Austria

Reviewed by: Ruth D. Fritsch-Stork, Sigmund Freud University Vienna, Austria

Hanan Altaee, University of Babylon, Iraq