Seletalisib for Activated PI3Kδ Syndromes: Open-Label Phase 1b and Extension Studies

This article is featured in Top Reads, p.2939

Activated PI3Kδ syndrome (APDS) is a rare genetic form of primary immunodeficiency with diverse clinical manifestations resulting from overactivity of PI3Kδ signaling that affects both B and T cells. This subtype of primary immunodeficiency was first described in 2013 (1, 2), and currently over 200 cases of APDS have been identified worldwide (3, 4). Patients often present with recurrent respiratory infections in infancy or childhood followed by lymphoproliferation, immune dysregulation phenomena (in particular, autoimmune cytopenia and inflammatory bowel disease), bronchiectasis, and recurrent herpes virus infections, plus an increased risk of lymphoma (1, 2, 4–6). Abnormally high serum IgM and low serum IgG2 and IgA are commonly reported (1, 2, 7, 8), and the B and T cell profile is characterized by high numbers of transitional B cells, low numbers of memory B cells, and high numbers of senescent T cells (1, 2, 6–8).

Two different genetic causes of APDS have been described, resulting from mutations in the PIK3CD (APDS1) and PIK3R1 genes (APDS2) encoding the catalytic p110δ subunit and the regulatory p85α subunit of PI3Kδ, respectively (1, 2, 6, 8, 9). APDS1 and APDS2 share most clinical features, although neurologic and developmental manifestations appear to be predominantly associated with APDS2, and bronchiectasis is more frequent in APDS1 (3, 4, 6, 7).

As there is no approved targeted APDS treatment, current therapy includes antibiotic prophylaxis and Ig replacement therapy (IGRT) to manage recurrent infections and hypogammaglobulinemia, and steroids and other immunosuppression drugs for immune dysregulation (4, 6, 7). Treatment options include the mammalian target of rapamycin (mTOR) inhibitor rapamycin, which acts on the PI3K/protein kinase B (AKT)/mTOR signaling pathway (2, 6, 7, 10), and was recently shown to reduce lymphoproliferation in 26 patients, with lower responses in cytopenia and bowel inflammation (4). A case of a young, female patient with APDS2 has suggested that theophylline treatment could reduce lymphoproliferation, improve immunologic abnormalities, and reduce infections requiring antibiotics (11). To date, hematopoietic stem cell transplantation (HSCT) is the only definitive curative option. Two series of patients treated with HSCT reported survival rates of 9 out of 11 and 7 out of 9 patients, improvements in most clinical symptoms, and significant improvements in humoral immunity (12, 13). To address the PI3Kδ overactivity, several PI3Kδ inhibitors are in development for APDS1 but not APDS2, including leniolisib, for which initial results are encouraging (14), and nemiralisib (15).

Seletalisib is a potent and selective oral PI3Kδ inhibitor with potential for patients with APDS (16–18). We assessed the ability of seletalisib to block PI3K signaling in T cell blast cultures from patients with APDS1 and APDS2. We subsequently conducted this phase 1b open-label study (European Clinical Trials Database 2015-002900-10) followed by an extension study (European Clinical Trials Database 2015-005541) to assess the safety, tolerability, pharmacokinetics, and efficacy of seletalisib in patients with APDS1 and APDS2.

The effect of seletalisib on the expression of AKT phosphorylated at Ser473 (pAKT^S473) was determined by flow cytometry in T cell blast cultures generated from two healthy donors (female aged 55 y and an unknown blood bank sample), three patients with APDS1 (female aged 8 y and two males aged 15 and 16 y), and three patients with APDS2 (female aged 6 y and two males aged 5 and 22 y). Details are given in the Supplemental Table I.

The phase 1b, open-label, multicenter, exploratory study aimed to assess the safety, tolerability, pharmacokinetics, and efficacy of seletalisib in patients with APDS. Patients were recruited in four centers in France, Germany, Italy, and Spain. This study comprised a screening period of up to 19 d and a 12-wk treatment period. Patients who completed the study could enter an extension study. Screening for the extension took place at week 12 of the phase 1b study, with treatment reinitiated ≤14 d thereafter. Details are supplied in Supplemental Table I.

Both studies were conducted in accordance with the Declaration of Helsinki, and the protocols were approved by the relevant ethics committees. All patients provided written informed consent prior to entry into each study.

The inclusion criteria were as follows: patients (≥12 y of age) with a confirmed genetic diagnosis of APDS and currently active clinical manifestations related to APDS were eligible. For the extension study, patients were required to have completed the phase 1b study and be expected to gain reasonable benefit from seletalisib treatment.

The exclusion criteria were as follows: treatment with mTOR inhibitors within the previous 6 wk, or with cyclosporine; history of clinically significant gastrointestinal conditions considered incompatible with the study in the previous 12 mo (including inflammatory bowel disease, peptic ulcers, or recurrent colitis with rectal bleeding); positive test for HIV, hepatitis B or C; severe infection during screening; history of allogeneic bone marrow transplantation; known latent or high risk of tuberculosis infection; high levels of alanine aminotransferase (ALT), aspartate aminotransferase, alkaline phosphatase or total bilirubin; or history of, or known risk factors for, liver disease. Patients with no clinical manifestations or only advanced bronchiectasis in the previous 12 mo, body weight ≤20 kg, current or previous lymphoma, low WBCs, or a low absolute neutrophil count were also excluded. Patients with new significant uncontrolled condition or ongoing drug-related serious treatment–emergent adverse events (SAE) from the phase 1b study were excluded from the extension study.

In the phase 1b study, all patients received oral seletalisib in a 15–25 mg dose once daily (based on body weight ≥20–≤35 kg, 15 mg; >35–≤50 kg, 20 mg; >50 kg, 25 mg). Dose adjustments could be made up to a maximum of 45 mg once daily.

In the extension study, patients initially received the same dose as at the end of the phase 1b study. Dose adjustments were permitted.

In the phase 1b study, the primary end point was safety and tolerability, including incidence and type of adverse events (see Supplemental Table I for definitions). The secondary end point was plasma concentration of seletalisib up to 12 wk.

Exploratory efficacy endpoints included changes in clinical features and disease activity. Details of the exploratory efficacy endpoints and the exploratory immunology and pharmacodynamic endpoints are given in Supplemental Table I.

The extension study had the same endpoints as the phase 1b study, with the exception of responder status as assessed by the investigator. The results for the extension study are reported after the phase 1b results.

No formal sample size calculation was performed. Descriptive statistics were presented for safety and efficacy data in both studies. Details and definitions of the analysis sets are provided in Supplemental Table I.

Because of the small sample size in this trial, individual patient data cannot be adequately anonymized, and there is a reasonable likelihood that individual participants could be reidentified. For this reason, data from this trial cannot be shared.

Seletalisib potently inhibited PI3K signaling, measured by the reduction in the proportion of pAKT⁺ cells among T cell blasts generated from PBLs of patients with APDS1 and APDS2 (Fig. 1, Supplemental Fig. 1). Inhibition occurred both in the absence and presence of T cell activation induced by anti-CD3 stimulation. The ranges of IC₅₀ values were similar for T cell blasts from patients with APDS1 and APDS2. The geometric mean (range) IC₅₀ was 8 (3–15) nM for APDS1 and 13 (8–20) nM for APDS2 in the absence of anti-CD3 (in healthy donors the pAKT signal was too low to calculate reliable IC₅₀ data). In the presence of anti-CD3, geometric mean (range) IC₅₀ was 17 (9–42) nM for healthy donors, 21 (7–50) nM for APDS1, and 28 (20–33) nM for APDS2.

In the phase 1b study, seven patients received seletalisib, of whom five completed the study. Two patients discontinued because of treatment-emergent adverse events (TEAEs), one after 16 d and one after 42 d of treatment.

Four patients entered the extension study and received seletalisib. Three patients completed ≥84 wk of treatment in the extension study; one patient withdrew consent at week 36 for personal reasons.

Most patients were adolescents; median age was 15 y in the phase 1b study and 13.5 y in the extension study (range 12–24 y for both studies). Three patients had APDS1, and four patients had APDS2; two patients with APDS1 and two with APDS2 entered the extension study. The patients had heterogeneous disease, heterogeneous clinical presentations (Table I), and baseline immunological profiles (Table II). Most patients had CD19⁺ B cell lymphopenia, and some had low CD4⁺ T cell counts with corresponding high CD8⁺ T cell counts. Most patients had high IgM and low IgA serum levels.

Duration of seletalisib exposure for individual patients is given in Supplemental Fig. 2.

In the phase 1b study, geometric mean plasma trough concentrations of seletalisib over time demonstrated that patient exposure to seletalisib was as expected. Plasma concentrations by individual patient are presented in Fig. 2.

The majority of patients demonstrated an improvement in at least one clinical feature during the phase 1b study that persisted over time in patients who went on to participate in the extension study. In peripheral lymphadenopathy assessments, two patients had partial remission of lesions by week 12; two discontinued before week 12 and could not be assessed. Change from baseline at week 12 of the sum of the product of the diameters was −67.6% in one patient and −80.6% in another. Among the four patients assessed for mediastinal and/or intra-abdominal lymphadenopathy, all had stable disease. One patient had an improvement in lung function from baseline at week 12 as measured by increased forced expiratory volume in 1 s (0.76 l; 51.4%), decreased forced vital capacity (−0.41 l; −15.4%), increased forced expiratory volume in 1 s/forced vital capacity ratio (0.44 l; 80.0%), and decreased diffusing capacity of the lung for carbon monoxide (−5.83 ml/min/mmHg; −17.0%). One patient experienced normalization of thrombocytopenia and another had normalization in chronic enteropathy.

During the extension study, the reduction in lymphoproliferation was maintained in one of two patients with data available. Lung function remained stable in the three patients assessed, and the normalization of thrombocytopenia in the phase 1b study was maintained. No patients had colitis during the extension study.

At the end of the phase 1b study, four of seven patients (57.1%) were classified as responders according to the investigator; one of seven patients (14.3%) remained stable during the study, without any significant benefit, and was judged as a nonresponder by the investigator. Two patients could not be assessed because they discontinued the study before week 12.

During the phase 1b study, seletalisib treatment was associated with a trend for decreased disease activity as measured by the physician’s global assessment of disease activity (PhGADA) and patient/caregiver’s global assessment of disease activity (PtGADA), with a mean (SD) change from baseline at week 12 of −7.6 (21.8) and −10.1 (18.5), respectively. Individual patient data indicated that disease activity varied among patients (Fig. 3).

No overall trends in PhGADA and PtGADA were observed during the extension study. Disease activity improved from baseline in two patients (decreased PhGADA and PtGADA) at most visits and worsened in two patients (increased PhGADA and PtGADA) at all visits.

At baseline, there was an increased percentage of transitional B cells (CD19⁺/CD21⁺CD24⁺⁺) and a decreased percentage of naive B cells (CD19⁺/CD21⁺⁺). In the phase 1b study, following treatment with seletalisib, there was a reduction in transitional B cells (CD19⁺/CD21⁺CD24⁺⁺; Fig. 4A) and a corresponding increase in naive B cells (CD19⁺/CD21⁺⁺; Fig. 4B). No other B cell subpopulations demonstrated clear changes. There were no substantial changes in CD4⁺ or CD8⁺ T cells during the study; however, small reductions in senescent effector T cells (CD3⁺CD8⁺/CD57⁺; Fig. 4C) and a trend toward a decrease in nonregulatory CD4⁺ (CD4⁺/CD25⁻CD127⁻ [mean: 51.9% at baseline and 37.2% at week 12]) and CD8⁺ (CD8⁺/CD25⁻CD127⁻ [mean: 79.1% at baseline and 65.4% at week 12]). T cells (data not shown) were observed. Most patients demonstrated a reduction in phospho-S6 ribosomal protein (p-S6) expression in unstimulated CD19⁺ B cells (Fig. 4D) and stimulated CD3⁺ T cells (Fig. 4E). In the extension study, these changes were largely sustained; all patients had decreased p-S6 in unstimulated CD19⁺ B cells and stimulated CD3⁺ T cells from baseline (Fig. 4).

In exploratory analyses of subpopulations of immune cells measured centrally, there were no clear or consistent changes in absolute cell numbers of CD19⁺ B cells, transitional B cells as measured by a different combination of markers (CD19⁺CD10⁺CD27⁻), CD4 T cells, and naive T cells in the phase 1b and extension studies. CD8 effector memory cells (CD3⁺CD8⁺CD45RA⁻CCR7⁻) had no clear or consistent changes in the phase 1b study but decreased in the extension study (Supplemental Table II).

No clear changes were observed in serum Igs (IgA, IgD, IgE, IgG, or IgM) following treatment with seletalisib in the phase 1b study (Supplemental Table III). Two patients in the extension study, both with APDS1, had normalization of elevated IgD and IgM, respectively (Supplemental Table III). It should be noted that all patients had current or previous IGRT.

Viral DNA was either not detected by PCR or values were at the lower limit of quantification at baseline. No clinically significant (>1 log) changes in EBV or CMV viral load were observed at week 12, or during the extension study.

All patients in the phase 1b study reported TEAEs during the study (Table III). The most common were infections (six out of seven patients); these were mild to moderate in severity and not considered to be drug related. The most frequent individual TEAEs were nasopharyngitis, pyrexia, rhinitis, sinusitis, aphthous ulcer, and headache (two patients each). No cases of rash, neutropenia, or diarrhea were reported, although one patient with pre-existing chronic enteropathy experienced colitis. Drug-related TEAEs were reported by four of the seven patients who had 12 events (Table III), most commonly aphthous ulcer (two events) and increased hepatic enzyme (two events). Two patients reported TEAEs leading to discontinuation. One discontinued because of increased hepatic enzyme, which was considered drug related, and a potential drug-induced liver injury. This patient had a history of elevated liver enzymes. The second patient discontinued because of an SAE of potential drug-induced liver injury of moderate intensity and had no history of elevated liver enzymes or liver damage. In total, three patients reported SAEs. The other SAEs were hospitalization (to enable restarting of seletalisib under observation), which was not considered to be seletalisib related, and severe colitis in another patient (as mentioned above), also not considered to be seletalisib related. There were no deaths during the study period.

All patients in the extension study experienced TEAEs. The most frequent category was infections (in all four patients), which were mild to moderate in severity and not considered to be seletalisib related. One patient reported a seletalisib-related TEAE of aphthous ulcer (four events; Table III). One patient reported an SAE of severe stomatitis, which was not considered to be seletalisib related and resolved after treatment; this patient also had episodes of nonserious aphthous ulcer (mild-to-moderate severity, not considered seletalisib related). No patients discontinued because of TEAEs in the extension study, and there were no deaths.

No notable, clinically relevant abnormalities in hematology, vital signs, and 12-lead electrocardiogram parameters were observed in the phase 1b study; urinalysis parameters were generally within normal ranges. TEAEs related to abnormal clinical chemistry were reported by four patients and included hyperglycemia, drug-induced liver injury, increased hepatic enzyme (two events), and hypertransaminasemia. In the extension study, one patient reported two TEAEs related to abnormal hematology (decreased hemoglobin and increased C-reactive protein) and one TEAE related to vital signs (decreased weight); these were not considered related to seletalisib, and none were clinically significant.

This open-label phase 1b study of seletalisib in patients with APDS1 or APDS2 demonstrated an acceptable safety profile, and exploratory analyses demonstrated positive effects on several clinical manifestations of APDS and immune cell dysfunction. Results from the extension study suggested that seletalisib had a favorable risk–benefit profile in the longer term. A broad range of patients with active APDS were enrolled in the study, resulting in a population with heterogeneous clinical features and both types of APDS. Target engagement was demonstrated ex vivo by expression of pAKT⁺ in T cell blasts and in vivo by reductions from baseline in p-S6⁺ B and T cells.

Exposure to seletalisib was in line with that in studies with healthy volunteers (18). Five of the seven patients in this study tolerated seletalisib treatment during the phase 1b study. Of the four patients who entered the extension study, three patients tolerated seletalisib for the duration of the study (a further 84 wk, or 96 wk overall), and a fourth patient tolerated seletalisib for a further 36 wk (48 wk overall) until they withdrew consent for personal reasons.

The most common category of TEAEs in the phase 1b study and its extension was infections, most frequently nasopharyngitis, rhinitis and sinusitis. This was not unexpected, as respiratory infections are a frequent symptom of APDS (1, 2, 5, 6). The patient who discontinued the phase 1b study because of a TEAE of increased hepatic enzymes had high ALT and aspartate aminotransferase levels at screening and high ALT levels at baseline, together with a history of elevated liver function enzymes that were thought to be related to APDS. The patient who discontinued because of an SAE of potential drug-induced liver injury had no prior history of abnormal liver function tests, and liver enzymes were within the normal range at enrollment, although an underlying APDS-related liver disease was suspected. These TEAEs suggest that monitoring hepatic enzymes is important when treating patients with PI3Kδ inhibitors and that caution should be considered when prescribing seletalisib in patients with existing liver abnormalities (19). Among 33 patients with APDS, 30% had liver disturbances; elevation of liver transaminases was more frequent in patients with APDS2, and four of five liver biopsy samples suggested nodular regenerative hyperplasia (19). Seletalisib led to clinical improvement in the majority of patients in the phase 1b study, notably partial remission of peripheral lymphadenopathy (n = 2) that was sustained in the extension study in one of the two patients with data available, improvements in lung function (n = 1), normalized thrombocytopenia (n = 1), and normalized gastrointestinal status (n = 1). There was a trend for reduction in disease activity after 12 wk of seletalisib treatment, principally in PhGADA, and the investigators noted that subjective improvements were also reported by parents and patients in patient quality of life, concentration at school, and quality of sleep in four patients. The number of patients in the study was too small to draw any conclusions about the comparative efficacy of seletalisib in patients with APDS1 versus APDS2.

Changes in immune cell subpopulations suggested an effect towards normalization of immune function with seletalisib treatment, especially of B cell maturation, with a decrease in transitional B cells and a corresponding increase in naive B cells. Changes in transitional B cells, naive B cells, nonregulatory CD4⁺ and CD8⁺ T cells, and senescent CD8⁺ T cells were sustained with long-term seletalisib treatment. Serum Igs did not show any clear changes in the phase 1b study; the high baseline IgD of one patient and the high baseline IgM of another patient normalized during the extension study. The potential restoration of IgG production could not be meaningfully assessed because of continued IGRT during the study. Absence of complete restoration of the humoral compartment may have been due to previous damage to the lymph node architecture, which might suggest that early treatment could increase efficacy.

Seletalisib inhibition of PI3K signaling was demonstrated by monitoring pAKT^S473 in T cell blasts derived from peripheral blood of patients with APDS1 and APDS2. In the phase 1b study, p-S6⁺CD19⁺ B cells decreased from baseline to week 12, indicating that seletalisib inhibited PI3Kδ; this decrease was sustained during the extension study.

Other PI3Kδ inhibitors are in development for treating APDS. Results are available from a small open-label study of leniolisib (14), and an extension study is in progress (20). The leniolisib study population included patients with APDS1 only, and the population was generally older (aged 17–31 years) than that of the current study. Leniolisib doses increased over the 12-wk treatment period, whereas in the current phase 1b study, most patients had continuous treatment at the same dose, so the two studies may not be wholly comparable. Both studies observed reductions in lymphadenopathy and normalization of immune cell subpopulations. Additional endpoints examined in the current trial suggest that other clinical manifestations of APDS may also respond to seletalisib, such as the patient with marked improvement in respiratory function. Interestingly, efficacy has been noted in an APDS patient treated with theophylline (11), a drug used in severe asthma that appears to inhibit PI3Kδ (21). It remains to be seen whether this represents an advantage over treatment with rapamycin (4). A further phase 2 study in patients with APDS1 has recently been completed using the inhaled PI3Kδ inhibitor nemiralisib (15).

A preclinical study has suggested that the use of PI3Kδ inhibitors may increase genomic instability (22). This effect should be carefully considered, as such inhibitors would likely be administered to APDS patients for years. Regardless of treatment type, patients should be monitored in the long term for development of lymphoma (2, 3, 6), as patients with APDS have a predisposition for B cell lymphoma (3, 4, 6, 7, 23).

Limitations of the studies include the small number of patients treated and the open-label study designs. However, as APDS is a very rare, newly described disease, the numbers recruited were in line with feasibility and with other trials in this patient population (14).

To conclude, seletalisib demonstrated an acceptable safety profile and had positive effects on several clinical features and disease activity at the expected drug exposure in patients with APDS1 and APDS2. In addition, seletalisib improved several immune cell abnormalities in both APDS1 and APDS2. A favorable risk–benefit profile was maintained in the extension study for up to 84 wk of treatment. Seletalisib may be an attractive treatment option for this patient population, but results would need to be replicated in confirmatory studies with long-term follow-up. Additionally, it may be used as a bridge treatment prior to HSCT. Given the high unmet need in this serious disease, treatment with seletalisib may prove to be an appropriate and innovative option.

We thank the patients and caregivers, in addition to the investigators and teams, who contributed to this study. We thank Marco Cirillo, Roberta Lombardi, Giorgia Copponi (Bambino Gesù Children’s Hospital, Rome, Italy); Alberto Tommasini (Trieste, Italy); Antonino Trizzino (Palermo, Italy); Maria Chiriaco, (Bambino Gesù Children’s Hospital University of Rome, Italy); Henrike Ritterbusch and Ilka Fuchs (Medical Center – University of Freiburg, Freiburg, Germany); and Emma Jones and Ingrid Franklin (UCB Pharma) for contributions to the study; Loic Chentout and Marie-Céline Deau (Paris, France) for technical assistance, Nuria Murtra (Barcelona, Catalonia, Spain) for patient logistics, and Sucharita Shankar (UCB Pharma) for assistance with data analysis and interpretation of results. We thank the Clinical Investigation Center at Necker Children’s Hospital (Paris, France) and the Clinical Research Unit and the Advanced Diagnostic Unit at the Center for Chronic Immunodeficiency (Freiburg, Germany) for their contribution to the care of patients and for technical support. We acknowledge Sally Cotterill, Certified Medical Publication Professional of iMed Comms, an Ashfield Company, part of UDG Healthcare plc, for medical writing support that was funded by UCB Pharma in accordance with Good Publication Practice guidelines (http://www.ismpp.org/gpp3). We also acknowledge Linda Feighery, Certified Medical Publication Professional of UCB Pharma for publication and editorial support.

N.D. is an employee and shareholder of UCB Pharma. M.J. is an employee of UCB Pharma. P.S.-P. has received grants from CSL Behring, Gilead Sciences, Grifols S.A., Octapharma, and Takeda. He has received personal fees from CSL Behring, Gilead Sciences, Grifols S.A., and Takeda and nonfinancial support from CSL Behring, Gilead Sciences, Grifols S.A., Pfizer, and Takeda. A.P. was an employee of UCB Pharma at the time this study was conducted, is currently employed by Exscientia Limited, and holds share options for UCB Pharma. G.I.J. is an employee and shareholder of UCB Pharma and a shareholder of AstraZeneca and Pfizer. E.H. was an employee of UCB Pharma at the time the studies were conducted. He has been previously employed by Johnson & Johnson and Takeda and is a current employee of Galapagos. D.C. is an employee of UCB Pharma. J.M. was employed as a consultant for this study by UCB Pharma. D.Y. is an employee of UCB Pharma. S.E. has received a research grant from UCB Pharma and Sandoz. A.M.-N. has received personal fees from CSL Behring and Takeda and nonfinancial support from CSL Behring, Takeda, and Octapharma. M.C. is cofounder of Smart Immune, is a designated inventor on published patent application WO2017/198590, and has consulted for Cellectis. S.K. is a CNRS staff researcher. He reports grants and payments for service agreements and travel from UCB Pharma and is a designated inventor on published patent application WO2017/198590. The other authors have no conflicts of interest.