The survey was distributed to members of the ASPN and CARRA who completed fellowship programs in pediatric nephrology and pediatric rheumatology. The 76 respondents from ASPN and 41 respondents from CARRA represented approximately 15% of the eligible members from each organization. All respondents were board-eligible or board-certified in their subspecialty with the vast majority having at least two years of post-fellowship experience, and 45% of pediatric nephrologists and 60% of pediatric rheumatologists having greater than 10 years of experience. Fifty-two percent of pediatric nephrologists reported managing less than 25 patients with SLE, and half of pediatric rheumatologists reported managing 25 to 100 patients with SLE (Table 1). Only 51% of the pediatric nephrologists and 24% of the pediatric rheumatologists surveyed in this study follow a standard protocol for treatment of LN (Table 1) reflecting the lack of guidance from RCTs for pediatric LN.
For the case of refractory class IV LN (See Additional File 1), a 16-year-old female failed induction therapy with seven monthly doses of CYC 500–1000 mg/m2 IV in addition to steroids. Although fever, rash, and arthritis improved, a repeat renal biopsy showed persistent activity, little chronicity, and no evidence of membranous LN. In the first follow-up scenario (See Additional File 1), patient was found to have no renal response to induction therapy with hypertension, peripheral edema, hypocomplementemia, hypoalbuminemia, positive dsDNA antibody, persistent proteinuria, and active urine sediment, as well as a significant increase in serum creatinine. Survey respondents were asked to consider their next choice in immunosuppressive agent. Pediatric nephrologists and rheumatologists agreed with the top choice of MMF in combination with RTX (42% vs. 44%, Fig. 1). CYC in combination with RTX (16% vs. 27%, Fig. 1) and MMF alone (20% vs. 20%, Fig. 1) were the next most common choices for therapy. Overall, while the choices varied within groups, there was no statistically significant difference between nephrologist and rheumatologist responses (p = 0.40).
In the second follow-up scenario of refractory class IV LN after CYC induction therapy (See Additional File 1), the patient was found to have mild renal response with hypertension, peripheral edema, hypocomplementemia, hypoalbuminemia, positive dsDNA antibody, and persistent proteinuria, but improvement in serum creatinine and only mild hematuria. Pediatric nephrologists and rheumatologists chose MMF alone (41% vs. 37%, Fig. 2) and MMF in combination with RTX (15% vs. 34%, Fig. 2) as their top choices for next step in immunosuppressive therapy in this scenario. Overall, there was no statistically significant difference between nephrologist and rheumatologist responses (p = 0.10). However, the majority of rheumatologists (53%) chose therapies that involved RTX alone or in combination, compared to only 31% of nephrologists that chose RTX therapies (Fig. 2). There was a statistically significant difference between the groups of nephrologist and rheumatologist responses in choices of RTX-containing regimens versus choices without RTX (p = 0.03).
In a third follow-up scenario of the refractory class IV LN case (See Additional File 1), the patient was found to have moderate renal response with persistent hypertension, resolution of peripheral edema, hypocomplementemia, hypoalbuminemia, positive dsDNA antibody, active urine sediment, but greater than 50% improvement in serum creatinine and proteinuria. Whereas a majority of pediatric nephrologists chose MMF alone (59%, Fig. 3), rheumatologists were split over use of MMF alone (38%) or in combination with RTX (35%) as their top choices for next step in immunosuppressive therapy. Overall, there was a statistically significant difference between nephrologist and rheumatologist responses (p < 0.01), unlike in the prior follow-up scenarios. Additionally, rheumatologists again chose more therapies that involved RTX compared to nephrologists (59% vs 21%, p < 0.01, Fig. 3). Thus, pediatric rheumatologists were more likely to use RTX in this scenario of active urine sediment but improvement in other renal parameters, and thus may be more aggressive with use of RTX. In all follow-up scenarios, there was no difference between groups of nephrologist and rheumatologist choices in therapies when considering additional CYC versus MMF or other options.
The second clinical vignette presented a case of a 12-year-old patient with SLE and class IV LN who achieved complete renal remission after induction therapy with CYC in addition to steroids for proliferative LN according to published CTP (11) (See Additional File 2). She was transitioned to MMF 1000 mg twice daily for maintenance therapy and tapering prednisone dose. Three months later, she developed nephrotic syndrome, hematuria with active urine sediment, hypocomplementemia, and high titer dsDNA antibody, without change in serum creatinine. She was diagnosed with renal flare, and repeat renal biopsy showed class IV LN with high activity and low chronicity scores. Pediatric nephrologists chose CYC alone (22%) or increased dose of MMF to 1500 mg twice daily (22%), whereas pediatric rheumatologists chose CYC in combination with RTX (36%) or CYC alone (21%) as their top choices for therapy (Fig. 4).
If the second case used MMF as the induction agent instead of CYC at diagnosis (See Additional File 2), nephrologists chose increased dose of MMF to 1500 mg twice daily (39%) or CYC alone (25%) to treat renal flare (Fig. 5). Rheumatologists chose CYC alone (27%) or MMF in combination with RTX (24%) as their top choices for treatment of renal flare (Fig. 5).
In the third follow up scenario of renal flare after complete renal remission was achieved with CYC and steroids followed by maintenance MMF therapy, the patient was also found to have elevated serum creatinine, rapidly progressive glomerulonephritis, and class IV LN on repeat renal biopsy (See Additional File 2). Nephrologists and rheumatologists (81% vs. 82%, Fig. 6) agreed that CYC with or without RTX were the best therapeutic choices for renal flare in this scenario, although more rheumatologists compared to nephrologists (52% vs. 22%, Fig. 6) would use CYC in combination with RTX.
In all three follow up scenarios involving case of renal flare after achieving remission with induction therapy, there was a statistically significant difference between the groups of nephrologist and rheumatologist responses (p = 0.02, 0.02, 0.04 respectively; Figs. 4–6). Additionally, rheumatologists chose more therapies that involved RTX than nephrologists in all scenarios (p = 0.03, 0.02, < 0.01 respectively; Figs. 4–6). There was no difference between groups of nephrologist and rheumatologist choices in therapies when considering additional CYC versus MMF or other options. Thus, pediatric rheumatologists chose RTX in more situations of renal flare than pediatric nephrologists, similar to the first case of proliferative LN refractory to induction therapy. Belimumab was included as a treatment option in both cases, however was rarely chosen.