Long-term efficacy and safety of secukinumab in patients with moderate-to-severe hidradenitis suppurativa: week 104 results from the SUNSHINE and SUNRISE extension trial

Abstract

Background

The SUNSHINE and SUNRISE phase III trials demonstrated sustained clinical efficacy of secukinumab in patients with moderate-to-severe hidradenitis suppurativa (HS) through 52 weeks. Patients completing the core trials could enter a 4-year extension trial.

Objectives

To evaluate the long-term efficacy, safety/tolerability and maintenance of clinical response to secukinumab through week 104 in the extension trial.

Methods

Patients with a hidradenitis suppurativa (HS) clinical response (HiSCR) at week 52 of the core trials (extension trial baseline visit) entered a randomized withdrawal period. HiSCR responders receiving subcutaneous secukinumab 300 mg every 2 or 4 weeks (SECQ2W/SECQ4W) through week 52 in the core trials were randomized 2 : 1 to continue secukinumab (SECQ2W-R-Q2W or SECQ4W-R-Q4W) or receive placebo (SECQ2W-R-PBO or SECQ4W-R-PBO) through week 104. The primary endpoint was time to loss of response (LOR; newly defined for this trial) through week 104 in week 52 HiSCR responders (SECQ2W-R-Q2W vs. SECQ2W-R-PBO and SECQ4W-R-Q4W vs. SECQ4W-R-PBO). Time to LOR was tested at 1.25% (one-sided) for each comparison (one-sided familywise alpha of 2.5%) through week 104. If LOR was met, patients could remain in the trial on open-label secukinumab treatment. Additional endpoints included safety and HiSCR. The trial was registered with ClinicalTrials.gov (NCT04179175).

Results

Overall, 84.3% of patients who completed the core trials entered the extension trial; 55.9% were week 52 HiSCR responders. The primary endpoint was not met for either secukinumab dosing regimen. The estimated risk reduction for LOR was 13% (SECQ2W-R-Q2W vs. SECQ2W-R-PBO; one-sided P = 0.25) and 30% (SECQ4W-R-Q4W vs. SECQ4W-R-PBO; one-sided P = 0.04). The median time to LOR was numerically longer in the secukinumab arms vs. placebo {SECQ2W-R-Q2W [283 days; 95% confidence interval (CI) 176, –] vs. SECQ2W-R-PBO [239 days; 95% CI 120, –]; SECQ4W-R-Q4W [365 days 95% CI 225, –] vs. SECQ4W-R-PBO [171 days; 95% CI 113–337]}. In week 52 HiSCR responders reporting LOR, 44% (SECQ2W-R-Q2W), 58% (SECQ2W-R-PBO), 40% (SECQ4W-R-Q4W) and 34% (SECQ4W-R-PBO) were achieving HiSCR at the time of LOR. Overall, the safety of secukinumab was consistent with the core trials.

Conclusions

The primary endpoint of this trial was not met. HiSCR was maintained in many patients at the time of LOR. The safety of secukinumab was consistent with the previously characterized safety profile in the core trials.

Lay Summary

Hidradenitis suppurativa (or ‘HS’ for short) is a chronic skin condition. It causes painful recurrent boils and abscesses (also called lesions). The results from two large clinical trials showed that a drug called ‘secukinumab’ was effective in treating symptoms of HS for 1 year. It was also well tolerated by patients.

This study followed the two large trials. It tested how effective and safe secukinumab is in patients who received treatment for HS for up to 2 years. Some patients who took secukinumab in the original trials were switched to receive a placebo (a dummy drug). Others kept taking secukinumab. Loss of response (LOR) to secukinumab was the main outcome measured. Because there is no established definition of LOR in HS, one was created for this trial. It was defined based on an increase in lesions over time. Patients who took secukinumab in the original trials and switched to placebo in this trial experienced LOR quicker than patients who continuously took secukinumab. However, many patients who experienced LOR showed an improvement in their HS.

Our findings suggest that secukinumab was well tolerated by people with HS. No new safety issues were found, other than those already known.

Linked Article: Bettuzzi and Grolleau Br J Dermatol 2025; 192:566–567.

What is already known about this topic?

• Hidradenitis suppurativa (HS) is a difficult-to-treat chronic recurring inflammatory disease associated with a high burden and substantial impact on quality of life.

• Only three biologic treatments (adalimumab, secukinumab and bimekizumab) are currently approved for treating moderate-to-severe HS.

• Robust long-term data are limited.

• The SUNSHINE and SUNRISE phase III trials demonstrated sustained clinical efficacy of secukinumab in patients with moderate-to-severe HS.

What does this study add?

• The SUNSHINE and SUNRISE extension trial describes long-term treatment with secukinumab, including a randomized withdrawal period [for week 52 HS clinical response (HiSCR) responders in the core trials] and an open-label period (for week 52 HiSCR nonresponders).

• The primary endpoint (newly defined time to loss of response between secukinumab and placebo arms) was missed.

• However, consistent with core trials, clinical benefits with secukinumab treatment were observed and no new safety concerns were identified.

Managing patients with hidradenitis suppurativa (HS) is challenging owing to the disease’s complexity.^1–3 Therapeutic approaches include a combination of medical therapies and surgical interventions.^2–7 To date, only three biologic treatments have been approved for moderate-to-severe HS: adalimumab [tumour necrosis factor (TNF)-α] inhibitor;⁸ secukinumab [interleukin (IL)-17A inhibitor];^9,10 and bimekizumab (IL-17A and IL-17F inhibitor).¹¹

SUNSHINE and SUNRISE were identical phase III randomized clinical trials that reported sustained clinical efficacy and a favourable safety profile of secukinumab in patients with moderate-to-severe HS through 1 year.¹² Given the naturally occurring symptom fluctuations and disease unpredictability,^13,14 understanding the long-term maintenance of clinical responses and safety of biologics in patients with HS are important.

SUNSHINE and SUNRISE included a 4-year extension trial designed to evaluate the long-term efficacy, safety and ­tolerability in patients with HS treated with secukinumab. Here, the week 104 results from the extension trial are reported.

Materials and methods

Trial design and patients

This trial (NCT04179175) is a multicentre 4-year extension trial of the landmark SUNSHINE and SUNRISE phase III core trials.¹² Adults (aged ≥ 18 years) with moderate-to-severe HS for ≥ 1 year were eligible for core trial participation.¹² Previous treatment with TNF-α inhibitors or stable doses of selected antibiotics were also permitted.¹² Patients who received treatment through week 52 and who completed the core trials were eligible for inclusion in the extension trial (Appendix S1; see Supporting Information).

The week 52 visit of the core trials was the first/baseline extension trial visit. The extension trial design differed based on a patient’s HS clinical response (HiSCR) status (responder/nonresponder) at week 52 of the core trials (Figure 1). HiSCR was defined by a ≥ 50% decrease in abscess and inflammatory nodule (AN) count with no increase in the number of abscesses and/or draining tunnels at week 52 vs. the weighted average for each component of the HiSCR (abscesses, inflammatory nodules, draining tunnels) recorded at baseline and/or screening of the core trials (Appendix S1). There was no official screening period in the extension trial.

Figure 1

Design of the SUNSHINE and SUNRISE extension trial. ^a The core trials included a placebo arm between baseline and week 16; all patients in the core trials were receiving secukinumab between weeks 16 and 52.¹² EOT-1/2, end of treatment period 1/2; FU, follow-up; HiSCR, hidradenitis suppurativa clinical response; R, responders at week 52 of the core trials; NR, nonresponders at week 52 of the core trials; LOR, loss of response; RWP, randomized withdrawal period; OL, open label; OLP, open-label period; PI, principal investigator; Q2W, every 2 weeks; Q4W, every 4 weeks; SEC, secukinumab 300 mg; Wk, week.

Week 52 HiSCR responders

Patients who were week 52 HiSCR responders entered a double-blind randomized withdrawal period (RWP) from week 52 to end-of-treatment 1 [EOT-1; week 104 or loss of response (LOR) visit (if before week 104)].

As no standard definition of LOR exists, LOR was newly defined for this trial and occurred if the following two conditions were met: (i) a ≥ 50% increase in AN count at any visit; and (ii) an increase of ≥ 3 in the absolute AN count vs. the average of the previous three visits or the week 52 visit (whichever was lower). If a patient experienced a ≥ 30% increase in AN count and an increase of ≥ 2 in the absolute AN count, they were reassessed within 2–4 weeks. A further increase of ≥ 2 in AN count at the reassessment visit was also considered LOR. Of note, LOR is different from loss of HiSCR.

HiSCR responders receiving subcutaneous (SC) secukinumab 300 mg every 2 weeks (SECQ2W) through week 52 were randomized 2 : 1 to continue SC SECQ2W (SECQ2W-R-Q2W) or receive SC placebo (SECQ2W-R-PBO) through EOT-1. HiSCR responders receiving SC secukinumab 300 mg every 4 weeks (SECQ4W) through week 52 were randomized 2 : 1 to continue SC SECQ4W (SECQ4W-R-Q4W) or receive SC placebo (SECQ4W-R-PBO) through EOT-1.

Following RWP completion, week 52 HiSCR responders continued through to end-of-treatment 2 (EOT-2; week 260) on open-label treatment. If LOR occurred before week 104, patients could continue in the trial, moving directly to open-label treatment after reinduction (placebo arms) or dummy reinduction (secukinumab arms; see Appendix S1).

Week 52 HiSCR nonresponders

All week 52 HiSCR nonresponders could continue open-label treatment with SC SECQ2W in the extension trial through EOT-2 (SECQ2W-NR-Q2W/SECQ4W-NR-Q2W). Week 52 HiSCR nonresponder results are reported in Appendix S1 (see Supporting Information). All patients who completed EOT-2, and patients who discontinued prematurely, entered a voluntary 8-week treatment-free follow-up period.

Objectives

The primary objective of the extension trial was to demonstrate the efficacy of secukinumab in week 52 HiSCR responders, and any subsequent LOR through week 104, compared with placebo (SECQ2W-R-Q2W vs. SECQ2W-R-PBO and SECQ4W-R-Q4W vs. SECQ4W-R-PBO). Secondary and exploratory objectives included the long-term safety and tolerability of secukinumab, HiSCR, skin pain response [numerical rating scale (NRS) 30] and Dermatology Life Quality Index (DLQI) response (see Appendix S1 for more information).

Statistical analysis

See Appendix S1 for detailed information on statistical analyses. Briefly, the primary endpoint analysis was conducted on the full analysis set of HiSCR responders.

A stratified [by region (Europe; Asia-Pacific, Middle East and Africa combined with Japan; Latin America and Canada combined with the USA) and bodyweight category (< 90 kg, ≥ 90 kg)] log-rank test was performed to assess the treatment effect of the risk of reaching LOR. Separate tests were performed to compare SECQ2W-R-Q2W vs. SECQ2W-R-PBO and SECQ4W-R-Q4W vs. SECQ4W-R-PBO. Hazard ratios (HRs) and corresponding 95% confidence intervals (CIs) were estimated using a stratified (by region and bodyweight category) Cox proportional hazards regression model with the extension trial treatment arm as an explanatory variable. Analogously to the stratified log-rank tests, two separate stratified Cox models were fitted.

A 2.5% alpha level (one-sided) was used to control for the familywise type I error. Both hypotheses (SECQ2W-R-Q2W vs. SECQ2W-R-PBO and SECQ4W-R-Q4W vs. SECQ4W-R-PBO) were tested at alpha/2 = 1.25% (one-sided). If the null hypothesis was rejected for one dosing regimen, but not for the other, the alpha could be shifted to the other regimen and the null hypothesis could be retested at a 2.5% alpha (one-sided).¹⁵

Results

Patients

Overall, 1084 patients were included in the core trials. At week 16, patients receiving placebo were switched to SECQ2W or SECQ4W, whereas those receiving SECQ2W or SECQ4W remained on the same treatment through week 52. Overall, 830 patients completed week 52 (SECQ2W, n = 280; SECQ4W, n = 270; placebo-SECQ2W, n = 136; placebo-SECQ4W, n = 144),¹² and 84.3% (n = 700/830) of patients entered the extension trial [week 52 HiSCR responders (n = 391); week 52 HiSCR nonresponders (n = 308); mis-randomized (n = 1)].

Figure 2 depicts patient disposition throughout the extension trial. Overall, 93.9% (n = 367/391) of week 52 HiSCR responders completed EOT-1 (week 104 or met LOR). At the time of data cutoff (26 May 2023), 92.6% (n = 362/391) had entered the open-label period. Treatment discontinuation in the RWP was low (6.1%; n = 24/391) with patient decision (2.8%; n = 11/391) being the main reason cited. At data cutoff, 65.3% (n = 201/308) of HiSCR nonresponders were undergoing open-label treatment.

Figure 2

Patient disposition through week 104 in the SUNSHINE and SUNRISE extension trial. All patients rolling over from the core trials into the extension trial were on active treatment with either secukinumab 300 mg every 2 weeks (SECQ2W) or secukinumab 300 mg every 4 weeks (SECQ4W). Patients receiving SECQ2W who were week 52 hidradenitis suppurativa clinical response responders (HiSCR-R) at the end of the core trials were randomized to either continue receiving SECQ2W or withdrawal [received placebo (PBO)]. Patients receiving SECQ4W who were week 52 HiSCR-R at the end of the core trials were randomized to either continue receiving SECQ4W or withdrawal (received PBO). Patients continued in the randomized treatment period until end-of-treatment period 1 (EOT-1), week 104 or the loss of response (LOR) visit, whichever was recorded first. After EOT-1, patients could opt to continue in the trial under open-label (OL) treatment (see ‘Entered into OL period’) or discontinue. Patients receiving SECQ2W who were week 52 HiSCR nonresponders (HiSCR-NR) at the end of the core trials went directly into the OL treatment period, remaining on SECQ2W. Patients receiving SECQ4W who were week 52 HiSCR-NR at the end of the core trials went directly into the OL treatment period but were uptitrated to SECQ2W treatment. Blue boxes represent randomized withdrawal (RW) treatment; yellow boxes represent OL treatment. AE, adverse event; FU, follow-up; n, number of patients; R, responders at week 52 of the core trials; NR, nonresponders at week 52 of the core trials. ^aNot all patients who completed the core trials entered the extension study; ^bpatients meeting the LOR criteria before week 104 could enter directly into OL treatment; ^cat the time of data cutoff (26 May 2023).

Baseline demographic and disease characteristics

Table 1 shows the baseline demographics and disease characteristics of patients from the core trials who were included in the extension trial. Overall, the baseline demographics of included patients were similar to the core trials and were generally balanced across arms.¹² Compared with other arms, fewer patients in the SECQ4W-R-Q4W arm (RWP) had Hurley stage III vs. Hurley stage II disease and reported fewer draining tunnels at baseline.

Table 1

Baseline demographics and disease characteristics of patients from the core trials who were included in the extension trial

Characteristica	Randomized withdrawal period: Week 52 HiSCR-R				OL period: Week 52 HiSCR-NR
	SECQ2W-R-Q2W (n = 136)	SECQ2W-R-PBO (n = 71)	SECQ4W-R-Q4W (n = 121)	SECQ4W-R-PBO (n = 63)	SECQ2W-NR-Q2W (n = 151)	SECQ4W-NR-Q2W (n = 157)
Age (years), mean (SD)	35.7 (11.3)	34.8 (10.6)	35.4 (12.6)	36.0 (11.3)	37.3 (11.0)	37.1 (11.1)
Female sex	73 (53.7)	38 (53.5)	64 (52.9)	28 (44.4)	91 (60.3)	78 (49.7)
Race
White	113 (83.1)	51 (71.8)	90 (74.4)	48 (76.2)	115 (76.2)	111 (70.7)
Black or African American	8 (5.9)	6 (8.5)	11 (9.1)	7 (11.1)	6 (4.0)	14 (8.9)
Asian	12 (8.8)	11 (15.5)	13 (10.7)	6 (9.5)	24 (15.9)	28 (17.8)
Otherb	3 (2.2)	3 (4.2)	7 (5.8)	2 (3.2)	6 (4.0)	4 (2.5)
Weight (kg), mean (SD)	93.7 (24.7)	92.4 (23.7)	93.0 (23.6)	94.0 (20.6)	92.2 (22.9)	95.1 (22.5)
BMI (kg m–2), mean (SD)	32.0 (7.8)	31.6 (7.5)	31.8 (7.7)	31.7 (7.5)	32.0 (7.5)	32.7 (7.3)
Smoking status
Current	79 (58.1)	35 (49.3)	52 (43.0)	39 (61.9)	86 (57.0)	88 (56.1)
Former	15 (11.0)	12 (16.9)	20 (16.5)	6 (9.5)	24 (15.9)	29 (18.5)
Hurley stage
II	78 (57.4)	41 (57.7)	78 (64.5)	37 (58.7)	80 (53.0)	86 (54.8)
III	51 (37.5)	29 (40.8)	35 (28.9)	23 (36.5)	65 (43.0)	66 (42.0)
Time since HS diagnosis (years), mean (SD)c	8.7 (7.2)	7.2 (5.5)	8.0 (7.4)	7.2 (6.4)	7.6 (6.1)	7.6 (6.8)
AN count, mean (SD)	13.8 (10.5)	13.9 (8.5)	14.2 (8.8)	11.2 (7.0)	13.7 (9.8)	12.3 (8.5)
Draining tunnel count, mean (SD)	2.4 (3.1)	2.8 (3.6)	2.0 (2.8)	3.1 (3.9)	3.0 (3.8)	3.0 (3.2)
NRS/skin pain, mean (SD)	5.0 (2.2)d	5.6 (2.6)e	5.0 (2.6)f	4.8 (2.6)g	5.6 (2.5)h	5.2 (2.5)i
Previous exposure to systemic biologics	33 (24.3)	13 (18.3)	25 (20.7)	14 (22.2)	30 (19.9)	37 (23.6)
Previous exposure to systemic antibiotics	109 (80.1)	59 (83.1)	102 (84.3)	49 (77.8)	136 (90.1)	132 (84.1)
Previous surgery for HS	54 (39.7)	35 (49.3)	43 (35.5)	22 (34.9)	60 (39.7)	67 (42.7)

Data are presented as n (%) unless otherwise stated. AN, abscess and inflammatory nodule; BMI, body mass index; HiSCR, hidradenitis suppurativa clinical response; HS, hidradenitis suppurativa; NR, nonresponders at week 52 of the core trials; NRS, numerical rating scale; OL, open label; PBO, placebo; Q2W, every 2 weeks; Q4W, every 4 weeks; R, responders at week 52 of the core trials; SEC, secukinumab 300 mg; W, week. ^aAll baseline demographics and disease characteristics are from baseline of the core trials; ^b‘Other’ included ‘American Indian or Alaska Native’ and ‘Multiple’; ^cthe date of first diagnosis collected in the core trials is evaluated with respect to the extension trial informed consent signature; ^dn = 127; ^en = 65; ^fn = 109; ^gn = 58; ^hn = 141; ⁱn = 145.

Primary endpoint analysis: time to loss of response

The extension trial primary endpoint – time to LOR – was not met for either secukinumab dosing regimen vs. placebo (Table 2). The estimated risk reduction for LOR for SECQ2W-R-Q2W vs. SECQ2W-R-PBO was 13% (HR 0.87, 95% CI 0.59–1.29; one-sided P = 0.25). The estimated risk reduction for LOR for SECQ4W-R-Q4W vs. SECQ4W-R-PBO was 30% (HR 0.70, 95% CI 0.47–1.05; one-sided P = 0.04). Kaplan–Meier estimates for the median time to LOR were numerically longer for the two secukinumab arms vs. the respective placebo (Figure 3). The median time to LOR for SECQ2W-R-Q2W was 283 days (95% CI 176, –) vs. 239 days (95% CI 120, –) for SECQ2W-R-PBO (44-day difference). Moreover, the median time to LOR for SECQ4W-R-Q4W was 365 days (95% CI 225, –) vs. 171 days (95% CI 113–337) for SECQ4W-R-PBO (194-day difference). Approximately 25% of patients who met LOR did so within the first 12 weeks (SECQ2W-R-Q2W, 26.9%; SECQ2W-R-PBO, 23.9%; SECQ4W-R-Q4W, 24.2%; SECQ4W-R-PBO, 30.2%) (Figure S1; see Supporting Information).

Figure 3

Cumulative incidence rate of loss of response (LOR) in week 52 hidradenitis suppurativa (HS) clinical response (HiSCR) responders. Cumulative incidence curve of time to event (loss of response) from week 52 (day 1) through week 104 (up to day 400) in the SUNSHINE and SUNRISE extension trial in the SECQ2W-R-Q2W vs. SECQ2W-R-PBO treatment arms (left panel) and the SECQ4W-R-Q4W vs. SECQ4W-R-PBO treatment arms (right panel). These curves are defined as 1 minus the Kaplan–Meier estimates. Patients who did not experience LOR through week 104, those with ≥ 2 consecutive assessments missing and those who permanently discontinued the trial [except for discontinuation due to lack of efficacy or adverse event (AE)] were censored at the last recorded visit. Patients taking rescue medications or permanently discontinuing the trial due to lack of efficacy or AE were considered as meeting LOR. Intake of prohibited medications/treatment or COVID-19-related events (missing study drug or discontinuing treatment) were ignored. R, responders at week 52 of the core trials; PBO, placebo; Q2W, every 2 weeks; Q4W, every 4 weeks; SEC, secukinumab 300 mg.

Table 2

Primary endpoint analysis of time to loss of response (LOR) in week 52 hidradenitis suppurativa clinical response (HiSCR) responders

	n/N (%)	KM estimates		HR estimates		P-value
		Median (days)	95% CI	Estimate	95% CI
SECQ2W-R-Q2W	73/136 (53.7)	283	176, –	0.87	0.59–1.29	0.25
SECQ2W-R-PBO	41/71 (57.7)	239	120, –
SECQ4W-R-Q4W	60/121 (49.6)	365	225, –	0.70	0.47–1.05	0.04
SECQ4W-R-PBO	41/63 (65.1)	171	113–337

Stratified Cox proportional hazards regression models, with treatment arm as an explanatory variable, and stratified by region and bodyweight category (< 90 kg, ≥ 90 kg), were used to estimate treatment effects and 95% confidence intervals (CIs). A one-sided alpha of 0.0125 was used for statistical testing for each comparison of secukinumab 300 mg (SEC) with placebo (PBO). One-sided P-values are based on a log-rank test stratified by region and bodyweight category (< 90 kg, ≥ 90 kg). Patients who did not experience LOR through week 104, those with ≥ 2 consecutive assessments missing, and those who permanently discontinued the trial [except for discontinuation due to lack of efficacy or adverse event (AE)] were censored at the last recorded visit. Patients taking rescue medications or permanently discontinuing the trial due to lack of efficacy or AE were considered as meeting LOR. Intake of prohibited medications/treatment or COVID-19-related events (missing study drug or discontinuing treatment) were ignored. HR, hazard ratio; KM, Kaplan–Meier; Q2W, every 2 weeks; Q4W, every 4 weeks; R, responders at week 52 of the core trials.

Post hoc analysis: time to regain abscess and inflammatory nodule count status after meeting the loss of response endpoint

Among all patients who met the LOR endpoint, 187 were eligible for post hoc evaluation of ‘time to regain AN count status’ following entry to the open-label period; 64.7% (n = 121/187) regained their prior AN count status by week 104, relative to the reference used to declare LOR (week 52 or average of the previous three visits, whichever was lower). Across arms, 32.6% (n = 61/187) regained AN count status within the first 12 weeks after meeting LOR (Figure S1; see Supporting Information).

Secondary endpoint analysis: safety

The safety of secukinumab in the RWP (week 52 to EOT-1) and the entire study period (week 52 until data cutoff) are presented in Table 3 and Table S1 (see Supporting Information), respectively.

Table 3

Week 104 safety of secukinumab in the randomized withdrawal period in week 52 hidradenitis suppurativa clinical response (HiSCR) responders

	Randomized withdrawal period: Week 52 HiSCR responders
	SECQ2W-R-Q2W (n = 137)a	SECQ2W-R-PBO (n = 70)	SECQ4W-R-Q4W (n = 121)	SECQ4W-R-PBO (n = 63)
Any AE(s), n (%)	78 (56.9)	41 (58.6)	71 (58.7)	31 (49.2)
EAIR/100 PTY	154.1	160.7	143.9	130.8
Most common TEAEs by PT (≥ 5%)
COVID-19b, n (%)	11 (8.0)	7 (10.0)	11 (9.1)	7 (11.1)
EAIR/100 PTY	13.1	17.3	14.7	20.3
HS n (%)	7 (5.1)	4 (5.7)	5 (4.1)	3 (4.8)
EAIR/100 PTY	8.0	9.7	6.4	8.3
Nasopharyngitis, n (%)	5 (3.6)	2 (2.9)	6 (5.0)	2 (3.2)
EAIR/100 PTY	5.8	4.7	7.6	5.6
Eczema, n (%)	3 (2.2)	4 (5.7)	3 (2.5)	1 (1.6)
EAIR/100 PTY	3.4	9.4	3.9	2.7
Patients with serious or other significant events
Death, n (%)	0 (0.0)	0 (0.0)	0 (0.0)	0 (0.0)
SAE, n (%)	2 (1.5)	4 (5.7)	5 (4.1)	5 (7.9)
Discontinued due to AE, n (%)	2 (1.5)	1 (1.4)	0 (0.0)	1 (1.6)
Safety topics of interest
Infections and infestations (SOC), n (%)	47 (34.3)	20 (28.6)	40 (33.1)	21 (33.3)
EAIR/100 PTY	69.7	57.7	61.4	74.9
URTI (HLT), n (%)	10 (7.3)	3 (4.3)	13 (10.7)	5 (7.9)
EAIR/100 PTY	11.9	7.1	17.1	14.5
Fungal infectious disorders (HLGT), n (%)	6 (4.4)	4 (5.7)	2 (1.7)	2 (3.2)
EAIR/100 PTY	7.0	9.5	2.5	5.4
Candida infections (HLT), n (%)	1 (0.7)	3 (4.3)	1 (0.8)	1 (1.6)
EAIR/100 PTY	1.1	7.0	1.3	2.7
Hypersensitivity (SMQ) (narrow), n (%)	8 (5.8)	6 (8.6)	5 (4.1)	5 (7.9)
EAIR/100 PTY	9.3	14.4	6.5	14.0
Malignant or unspecified tumours (SMQ), n (%)	0 (0.0)	1 (1.4)	0 (0.0)	0 (0.0)
EAIR/100 PTY	0.0	2.3	0.0	0.0
MACE (NMQ), n (%)	0 (0.0)	0 (0.0)	0 (0.0)	0 (0.0)
EAIR/100 PTY	0.0	0.0	0.0	0.0
IBD (CMQ), n (%)	0 (0.0)	0 (0.0)	0 (0.0)	0 (0.0)
EAIR/100 PTY	0.0	0.0	0.0	0.0
Colitis ulcerative (PT), n (%)	0 (0.0)	0 (0.0)	0 (0.0)	0 (0.0)
EAIR/100 PTY	0.0	0.0	0.0	0.0
Crohn disease (PT), n (%)	0 (0.0)	0 (0.0)	0 (0.0)	0 (0.0)
EAIR/100 PTY	0.0	0.0	0.0	0.0
IBD (PT), n (%)	0 (0.0)	0 (0.0)	0 (0.0)	0 (0.0)
EAIR/100 PTY	0.0	0.0	0.0	0.0

AE, adverse event; CMQ, customized Medical Dictionary for Regulatory Activities (MedRA) queries; EAIR, exposure adjusted incidence rate; HLGT, high-level group terms; HLT, high-level term; HS, hidradenitis suppurativa; IBD, inflammatory bowel disease; MACE, major adverse cardiovascular event; NMQ, Novartis MedDRA query; PBO, placebo; PT, preferred term; PTY, patient treatment-years; Q2W, every 2 weeks; Q4W, every 4 weeks; SAE, serious adverse event; R, responders at week 52 of the core trials; SEC, secukinumab 300 mg; SMQ, standardized MedDRA query; SOC, system organ class; TEAE, treatment-emergent AE; URTI, upper respiratory tract disease. ^aA serious Good Clinical Practice violation in the core trial was reported for one patient in this arm; they were subsequently excluded from the full analysis set/efficacy analyses but were included in the safety analyses. ^bThe study was conducted during the COVID-19 pandemic.

Adverse event (AE) reporting was balanced between arms in the RWP; the proportion [exposure-adjusted incidence rates/100 patient treatment-years (EAIR/100 PTY)] of any AEs between weeks 52 and 104 was 56.9% (154.1), 58.6% (160.7), 58.7% (143.9) and 49.2% (130.8) for SECQ2W-R-Q2W, SECQ2W-R-PBO, SECQ4W-R-Q4W and SECQ4W-R-PBO, respectively. Further, based on EAIR, no differences were observed between secukinumab dosing regimens in the entire study period up to data cutoff [any SECQ2W: 76.3% (129.3); any SECQ4W: 56.7% (138.4)], although these data should be interpreted with caution owing to differences in patient numbers between the any SECQ2W (n = 637) and any SECQ4W (n = 180) arms. The most common AEs were COVID-19, hidradenitis and nasopharyngitis, with no trend observed between treatment arms or secukinumab doses.

No deaths were reported in the extension trial up to data cutoff. Overall, discontinuation due to AEs/serious AEs (SAEs) occurred infrequently (Appendix S1; see Supporting Information).

Exploratory/post hoc analyses

Hidradenitis suppurativa clinical response

Of patients meeting LOR, 44% (n = 32/73), 58% (n = 23/40), 40% (n = 23/58) and 34% (n = 14/41) in the SECQ2W-R-Q2W, SECQ2W-R-PBO, SECQ4W-R-Q4W and SECQ4W-R-PBO arms, respectively, were achieving HiSCR at that time (Figure 4a). For patients in the SECQ2W-R-Q2W, SECQ2W-R-PBO, SECQ4W-R-Q4W and SECQ4W-R-PBO arms not meeting LOR, 88% (n = 46/52), 92% (n = 23/25), 92% (n = 49/53) and 95% (n = 20/21) respectively, were achieving HiSCR at week 104 (Figure 4a).

Figure 4

Exploratory efficacy analysis of hidradenitis suppurativa (HS) clinical response (HiSCR) through week 104 in week 52 HiSCR responders (HiSCR-R). (a) Bar graph demonstrating the proportion of patients achieving HiSCR at end-of-treatment period 1 (EOT-1; i.e. LOR or week 104). Error bars represent 95% confidence intervals. (b) Bar graph demonstrating absolute abscess and inflammatory nodule (AN) count change from baseline of the core trials in week 52 HiSCR-R at EOT-1. Error bars represent the SE. (c) Line graph demonstrating the proportion of patients achieving HiSCR in those receiving continuous secukinumab only from weeks 52 to 104. In (a) and (b), if a patient met LOR at the week 104 visit, they were counted both for LOR and week 104 groups. In (c), visits across both randomized withdrawal and open-label periods are considered and patients in the SECQ4W-R-Q4W arm who uptitrated to SECQ2W upon meeting the LOR criteria are included in the analysis. In (b), HiSCR-R at week 52 of the core trials did not equal 100% given the difference in definitions between identification of HiSCR-R at week 52 and calculation of HiSCR over time. The identification of HiSCR responders at week 52 was based on nominal visit week 52 and was computed with respect to the weighted average of lesions at core trial baseline and/or screening (see Appendix S1). The definition of HiSCR over time in the extension trial was based on the analysis visit and was computed with respect to the core trials baseline. All data are reported as observed. n, number of patients with available data at each timepoint; PBO, placebo; Q2W, every 2 weeks; Q4W, every 4 weeks; R, responders at week 52 of core trial; SEC, secukinumab 300 mg; SE, standard error.

Figure 4(b) details the post hoc analysis of the absolute change in AN count from baseline. Despite experiencing LOR, week 52 HiSCR responders reported a high mean (SE) change in AN count vs. baseline of the core trials [SECQ2W-R-Q2W –7.3 (1.14); SECQ2W-R-PBO –8.4 (1.15); SECQ4W-R-Q4W –9.4 (1.06); SECQ4W-R-PBO –4.8 (0.84)].

A further post hoc analysis of week 52 HiSCR responders on continuous secukinumab treatment from weeks 52 to 104 (including visits from the RWP and open-label treatment once a patient met LOR) is shown in Figure 4(c). HiSCR was sustained through week 104 in patients receiving continuous SECQ2W [SECQ2W-R-Q2W, week 52 (95.1%) vs. week 104 (76.5%)] and SECQ4W [SECQ4W-R-Q4W (including those who uptitrated to open-label SECQ2W upon meeting LOR), week 52 (98.1%) vs. week 104 (87.5%)].

Numerical rating scale 30 and Dermatology Life Quality Index responses

Figure 5 shows NRS30 and DLQI responses in week 52 HiSCR responders. Of patients meeting LOR, 34% (n = 10/29), 40% (n = 8/20), 35% (n = 8/23) and 33% (n = 6/18) in the SECQ2W-R-Q2W, SECQ2W-R-PBO, SECQ4W-R-Q4W and SECQ4W-R-PBO arms were NRS30 responders at that time, respectively (Figure 5a). For patients in the SECQ2W-R-Q2W, SECQ2W-R-PBO, SECQ4W-R-Q4W and SECQ4W-R-PBO arms not meeting LOR, 73% (n = 16/22), 50% (n = 3/6), 45% (n = 10/22) and 67% (n = 6/9) were NRS30 responders at week 104, respectively (Figure 5a). Further, NRS30 response was sustained in patients (including visits from the RWP and open-label treatment once a patient met LOR) on either continuous SECQ2W [SECQ2W-R-Q2W, week 52 (61.5%) vs. week 104 (58.2%)] or continuous SECQ4W [SECQ4W-R-Q4W, week 52 (62.3%) vs. week 104 (50.0%)] (Figure 5c).

Figure 5

Exploratory efficacy analysis of skin pain response/numerical rating scale (NRS) 30 and Dermatology Life Quality Index (DLQI) response through week 104 in week 52 hidradenitis suppurativa clinical response (HiSCR) responders. Bar graphs demonstrating the proportion of patients achieving (a) a skin pain response/NRS30 and (b) a DLQI response at end-of-treatment period 1 [EOT-1; i.e. at loss of response (LOR) and week 104]. Error bars represent 95% confidence intervals. Line graphs demonstrating the proportion of patients achieving (c) a skin pain response/NRS30 and (d) a DLQI response in patients receiving continuous secukinumab only from weeks 52 to 104. In (a) and (b), if a patient met LOR at the week 104 visit, they were counted both for LOR and week 104 groups. In (c) and (d), visits from both the randomized withdrawal and open-label periods are considered. Patients in the SECQ4W-R-Q4W arm who uptitrated to open-label SECQ2W upon meeting the LOR criteria are also included in the analysis. For the skin pain response/NRS30 analyses, only patients with an NRS ≥ 3 at baseline of the core trials were included. All data are reported as observed. n, number of patients with available data at each timepoint; PBO, placebo; Q2W, every 2 weeks; Q4W, every 4 weeks; R, responders at week 52 of the core trials; SEC, secukinumab 300 mg.

Of patients meeting LOR, 39% (n = 23/59), 39% (n = 13/33), 38% (n = 19/50) and 29% (n = 10/34) in the SECQ2W-R-Q2W, SECQ2W-R-PBO, SECQ4W-R-Q4W and SECQ4W-R-PBO arms, respectively, were DLQI responders at that time (Figure 5b). For patients in the SECQ2W-R-Q2W, SECQ2W-R-PBO, SECQ4W-R-Q4W and SECQ4W-R-PBO arms not meeting LOR, 70% (n = 33/47), 61% (n = 14/23), 54% (n = 27/50) and 59% (n = 10/17), respectively, were DLQI responders at week 104 (Figure 5b). Further, the proportion of patients (including visits from the RWP and open-label treatment once a patient met LOR) achieving a DLQI response was sustained in patients on either continuous SECQ2W [SECQ2W-R-Q2W, week 52 (61.7%) vs. week 104 (60.4%)] or continuous SECQ4W [SECQ4W-R-Q4W, week 52 (49.5%) vs. week 104 (54.9%)] (Figure 5d).

HiSCR, NRS30 and DLQI response in week 52 HiSCR nonresponders (open-label period) are described in Appendix S1 and Figure S3 (see Supporting Information).

Discussion

The results of the SUNSHINE and SUNRISE core trials validated the role of anti-IL-17A in HS management, demonstrating that secukinumab is effective in rapidly improving the signs and symptoms of moderate-to-severe HS, with a favourable safety profile and sustained responses through 1 year of treatment.¹² Subsequently, secukinumab received regulatory approval, making it the second approved biologic treatment for moderate-to-severe HS.^9,10 This extension trial aims to assess long-term efficacy and safety of secukinumab in moderate-to-severe HS, including under conditions of drug withdrawal.

The predefined primary endpoint of this extension trial (LOR in the RWP) was not met; there were no statistically significant differences in time to LOR in the secukinumab arms (SECQ2W-R-Q2W; SECQ4W-R-Q4W) vs. the respective placebo arms (SECQ2W-R-PBO; SECQ4W-R-PBO). Although not statistically significant, the median time to LOR was numerically longer for patients in the secukinumab arms (both SECQ2W and SECQ4W) vs. those receiving placebo. Particularly, the median time to LOR for patients in the SECQ4W-R-PBO arm was approximately 6 months shorter than for those in the SECQ4W-R-Q4W arm. Of note, patients receiving SECQ4W reported a longer time to LOR vs. SECQ2W. Similarly, in SUNRISE, SECQ4W had a more robust response in the placebo-controlled period. One theory for this observation was the differences in baseline characteristics; patients in the SECQ4W arm of SUNRISE had less severe disease and therefore may have been more responsive to treatment, consistent with other research showing that patients with a longer delay in diagnosis have poorer responses to biologic treatment,¹⁶ and that a longer delay in diagnosis likely results in more severe disease.^17,18 In this trial, more patients in the SECQ4W-R-Q4W arm had Hurley stage II vs. Hurley stage III disease, as well as a shorter disease duration than patients in the SECQ2W-R-Q2W arm, potentially contributing to this arm being more responsive to treatment.

LOR analyses are common in trials for biologic treatments and frequently requested by regulatory authorities to understand whether continuous treatment over time is required. However, to date, there is no consensus definition of LOR in the HS community. A previous post hoc analysis of the PIONEER open-label extension trial of adalimumab in moderate-to-severe HS used loss of HiSCR to assess loss of clinical response,¹⁹ while a German HS registry analysis used the more recently developed international hidradenitis suppurativa severity score system (IHS4).²⁰ In this trial, the definition of LOR used may have been too sensitive and represents a key limitation. The reference visit for detecting LOR was patient-specific and not fixed in time; it was at a timepoint where patients were managing symptoms well, with a low AN count vs. their core trial baseline values, making it easy to trigger LOR criteria, even with small increases in AN. This observation was further enhanced by the regression to the mean phenomenon, whereby any extreme variable tends to return closer to the average at subsequent measurements.²¹ These limitations are supported by the observation that approximately 25% of patients met LOR within the first 12 weeks of the extension trial. Meaningful reductions in AN count vs. baseline of the core trials were observed at the time of LOR, indicating that patients did not revert to baseline disease severity levels. Additionally, ­rollover to the extension trial was high; almost 85% of patients completing the core trials continued in the extension trial. A substantial proportion of patients remained in the trial through week 104, potentially indicating that patients were satisfied with their treatment.

Furthermore, a time-to-event analysis may not be clinically meaningful for a chronic disease with an inherent waxing and waning clinical course. LOR comparisons were affected by clinical benefits observed in patients receiving placebo, a phenomenon that is widely reported in HS,^22,23 and potentially explained by the naturally occurring fluctuation in symptoms of HS,¹³ with AN counts likely to vary over time naturally. Previous investigations have reported that variability in AN count in untreated patients with HS significantly contributes to placebo response rates.¹⁴ The post hoc analyses support the theory of natural fluctuations in AN count, as two-thirds of patients meeting LOR eventually regained their AN count status by week 104 following entry into the open-label period, with one-third doing so within the first 12 weeks. The complexity of the study design also resulted in difficulties interpreting the true clinical impact over the RWP. In fact, as per the study design, when patients met LOR, they were switched from placebo to secukinumab, making comparisons and trends between arms difficult. Lastly, it is important to note that the placebo arms were not a true placebo/biologic-naïve group as patients originally assigned to placebo at baseline received secukinumab between weeks 16 and 52 of the core trials and entered the extension trial without any washout or treatment-free period. Thus, there may have been some residual effects of secukinumab in these arms at the beginning of the RWP.

While secukinumab was not significantly different to placebo in terms of the protocol-specified time to LOR, overall, approximately 40% of patients treated with either SECQ2W or SECQ4W maintained HiSCR at the time of LOR, highlighting sustained clinical benefit. Further, of patients receiving continuous secukinumab (including patients reporting LOR), 76.5% of patients in the SECQ2W arm and 87.5% of patients in the SECQ4W arm (including patients uptitrating from SECQ4W to SECQ2W) continued to achieve HiSCR at week 104. Additional outcomes including improvement in skin pain response (NRS30) and quality of life (QoL; DLQI response) remained high in patients receiving continuous secukinumab over 2 years. The importance of improvement in skin pain and QoL in HS has been highlighted extensively in the literature, with both being recommended as core outcomes in HS clinical trials.²⁴

Safety observed in the extension trial was consistent with the core trials,¹² and similar to the known safety profile of secukinumab across indications.²⁵ No deaths were reported during the extension trial up to data cutoff, and SAE frequency was low and occurred at similar rates between arms. There were no dose-dependent findings except for fungal infections, which were higher in patients treated with SECQ2W, a finding similar to that reported in the core trials.¹² Overall, safety analyses confirmed the positive risk–benefit ratio associated with treatment with secukinumab.

Conclusion

The predefined primary endpoint of time to LOR through week 104 was not met for either secukinumab dosing regimen vs. placebo. Despite this, the median time to LOR was numerically longer for both secukinumab arms vs. the respective placebo arms and clinical benefit using other measures could be observed in patients who met LOR. A long-term sustained HiSCR was observed in week 52 HiSCR responders continuously exposed to secukinumab, supporting continued treatment with secukinumab. Secukinumab safety was consistent with the known profile of secukinumab across other indications and in patients with HS from the core trials. This confirms that secukinumab is a reliable treatment option for patients with moderate-to-severe HS in the long term.

Acknowledgements

All authors participated in the development of the manuscript. The authors thank Trudy McGarry PhD and Philip O’Gorman PhD (Novartis Ireland Limited, Dublin, Ireland) for editorial and medical writing support, which was funded by Novartis Pharma AG, Basel, Switzerland, in accordance with Good Publication Practice (GPP 2022) guidelines (http://www.ismpp.org/gpp-2022). Special thanks to all the patients with hidradenitis suppurativa (HS) who agreed to participate in this extension trial in an attempt to improve the knowledge and the care of HS. Further, the authors would like to thank all the study team members and investigators who contributed to the conduct of this trial.

Funding sources

This study was funded by Novartis Pharma AG, Basel, Switzerland.

Data availability

Novartis is committed to sharing, with qualified external researchers, access to patient-level data and supporting clinical documents from eligible trials. These requests are reviewed and approved by an independent review panel on the basis of scientific merit. All data provided are anonymized to respect the privacy of patients who have participated in the trials in line with applicable laws and regulations.

Ethics statement

The trial protocol and all amendments for the SUNSHINE and SUNRISE extension trial were reviewed by the Independent Ethics Committee or Institutional Review Board of each participating centre. The trial was done according to The International Conference on Harmonisation Guidelines for Good Clinical Practice that have their origin in the Declaration of Helsinki.

Patient consent

Written informed consent was obtained from each patient during the screening visit and before any study-specific procedure was done.

Supporting Information

Additional Supporting Information may be found in the online version of this article at the publisher's website.

Author notes

Conflicts of interest: A.B.K. reports grants from AbbVie, AnaptysBio, Aristea, Bristol Myers Squibb, ChemoCentryx, Lilly, Incyte, Janssen, MoonLake, Novartis, Pfizer, UCB Pharma and Sonoma Bio, and fellowship funding from AbbVie and Janssen paid to her institution; royalties from BIDMC; consulting fees from AbbVie, Alumis, Bayer, Bristol Myers Squibb, Boehringer Ingelheim, Eli Lilly, FIDE, Novartis, MoonLake, Janssen, Pfizer, Priovant, Sonoma Bio, Sanofi, Target RWE, UCB Pharma and Ventyx; stock in Ventyx; served on advisory boards for Target RWE; served as an advisory council member to the National Institute of Health Director; and serves on the board of directors of Almirall. F.G.B. received honoraria for participation in advisory boards, in clinical trials and/or as a speaker from AbbVie, ACELYRIN, Boehringer Ingelheim, Celltrion, Dr. Wolff, Incyte, InflaRx, Janssen-Cilag, Merck, Mölnlycke, MoonLake Immunotherapeutics, Novartis, Sanofi, Sitala and UCB Pharma. E.J.G.-B. has received honoraria from Abbott Products Operations, bioMérieux, GSK, UCB Pharma Sobi AB and ThermoFisher Brahms GmbH, and has received independent educational grants from Abbott Products Operations, AbbVie, bioMérieux, Johnson & Johnson, Incyte, MSD, Novartis, UCB Pharma, Sanofi and Sobi; and the Horizon 2020 European Grants ImmunoSep and RISCinCOVID and the Horizon Health grant EPIC-CROWN-2, POINT and Homi-Lung (granted to the Hellenic Institute for the Study of Sepsis). A.B.G. receives research/educational grants from Highlights Therapeutics, Bristol Myers Squibb, Janssen and UCB Pharma (all paid to Mount Sinai School of Medicine), and has received honoraria as an advisory board member and consultant for Amgen, AnaptysBio, Avotres Therapeutics, Boehringer Ingelheim, Bristol Myers Squibb, Dice Therapeutics, Eli Lilly, Highlights Therapeutics, Janssen, Novartis, Sanofi, Teva, UCB Pharma and XBiotech (stock options for rheumatoid arthritis). G.B.E.J. receives consultancy fees from Novartis and UCB Pharma; has received grants from AbbVie, ChemoCentryx, LEO Pharma, InflaRx, Novartis and UCB Pharma; and support for attending meetings from Sanofi. He is Vice President of the European Hidradenitis Suppurativa Foundation e.V. and on the board of the CHORD COUSIN Collaboration (C³) and CHORD; is co-coordinator of the ALLOCATE Skin group of the ERN (European Reference Network) Skin and vice chair of the Acne, Rosacea and HS (ARHS) Task Force group of the European Academy of Dermatology and Venereology. He is Editor-in-Chief of Dermatology. Z.R. has received honoraria (investigator/speaker or advisory boards) from AbbVie, Amgen, Boehringer Ingelheim, Incyte, Janssen, Lilly, Novartis, UCB Pharma and Sanofi. A.P.V. reports consulting fees from Janssen-Cilag and has received honoraria from AbbVie, Almirall, Bristol Myers Squibb, Janssen-Cilag, LEO Pharma, Lilly, MSD, Novartis and UCB Pharma. I.A., A. Bansal, F.G., R.M., B.P. and H.Z. are employed by Novartis. L.U. and S.R. were employed by Novartis at the time of the study and during manuscript preparation. A.A. serves as a board member for the Hidradenitis Suppurativa Foundation, as an investigator for Boehringer Ingelheim and Processa Pharmaceuticals, and as a consultant for AbbVie, Almirall, Boehringer Ingelheim, Incyte, InflaRx, LEO Pharma, Kymera, Novartis, Sanofi and UCB. A. Badat declares no conflicts of interest.