Abstract
Introduction
It remains unclear whether Janus kinase (JAK) inhibitors differ in efficacy and safety between elderly and non-elderly patients with ulcerative colitis.
Methods
We retrospectively compared outcomes between patients who started a JAK inhibitor at ≥65 years (elderly group) and those <65 years (non-elderly group).
Results
Among 228, 215, and 159 patients treated with upadacitinib, filgotinib, and tofacitinib, we identified 14, 36, and 13 elderly patients, respectively. There were no significant differences in efficacy between elderly and non-elderly patients for any of the three JAK inhibitors. The elderly group had a 3-fold higher risk of herpes zoster infection with upadacitinib or tofacitinib compared to the non-elderly group, whereas the risk with filgotinib was less than 3% in both groups. The non-elderly group had a 3-fold higher risk of acne with upadacitinib.
Conclusion
Adverse event risks with JAK inhibitors should be considered by age. Given the limitations of this study, including its retrospective design and small sample size, further studies with larger sample sizes are needed to validate our findings.
Keywords: Elderly, Janus kinase inhibitors, Ulcerative colitis
Introduction
The incidence of inflammatory bowel disease (IBD) is rising disproportionately, especially in the elderly [1]. While many biologics and Janus kinase (JAK) inhibitors have been approved for IBD, careful consideration is required when escalating therapy in elderly patients due to malnutrition, comorbidities, and frailty [2]. However, only a few studies have assessed the efficacy and safety of each advanced therapy in elderly IBD patients [3, 4].
In Japan, the three JAK inhibitors – filgotinib, tofacitinib, and upadacitinib – have been used to treat moderate-to-severe ulcerative colitis (UC) [5]. We previously conducted a multicenter study comparing the efficacy and safety of these three drugs. The results showed superior efficacy of upadacitinib compared to tofacitinib and filgotinib. While the risk of herpes zoster (HZ) was comparable across all three drugs, upadacitinib was significantly associated with an increased risk of acne [5, 6].
It remains unclear whether JAK inhibitors differ in efficacy and safety between elderly and non-elderly UC patients. Here, we aim to compare the efficacy and safety of the three JAK inhibitors in UC patients starting a JAK inhibitor before and after the age of 65 using our real-world data [5, 6].
Methods
This study was approved by the Ethics Committees of all participating institutions: University of Tsukuba (R04-289), Institute of Science Tokyo (M2022-285-01), Kitasato University School of Medicine (E2023022), the Jikei University School of Medicine (35-126 (11752)), Iwate Medical University (C-IRB2023-007), Juntendo University School of Medicine (C23-0038-H02), Okayama University Hospital (2306-021), Tokyo Yamate Medical Center (K-029), Ofuna Chuo Hospital (No. 2024-002), Kurume University School of Medicine (22586), and Tokyo Women’s Medical University (C24-026). The Ethics Committees approved the use of an opt-out consent approach and waived the requirement for informed consent because this study retrospectively analyzed existing clinical data.
We conducted a subanalysis of our multicenter, real-world study [5, 6] to compare the efficacy and safety of JAK inhibitors between patients starting a JAK inhibitor <65 years of age (non-elderly group) and patients starting a JAK inhibitor ≥65 years of age (elderly group). We assessed clinical outcomes as previously described [5, 6]. The primary outcome was clinical remission, defined as a partial Mayo score ≤1 with a rectal bleeding score of 0 or an Simple Clinical Colitis Activity Index (SCCAI) ≤2 with a blood in stool score of 0 [7, 8]. Secondary outcomes included clinical response (a ≥3-point reduction from the baseline SCCAI or a ≥2-point reduction from the baseline partial Mayo score) [7, 8], corticosteroid (CS)-free remission (clinical remission without systemic CS use), and endoscopic improvement (a Mayo endoscopic subscore of 0 or 1). Available follow-up data at 10 (±2), 26 (±2), or 58 (±4) weeks were assessed. In terms of clinical disease activity, partial Mayo scores [8] and SCCAI [7] were assessed and clinical disease activity was classified as mild (partial Mayo score 2–4 or SCCAI 3–5), moderate (partial Mayo score 5–7 or SCCAI 6–11), or severe (partial Mayo score >7 or SCCAI >11). Adverse events potentially related to JAK inhibitors, including acne, HZ, neutropenia, thrombosis, major adverse cardiovascular events (MACEs), malignancy, and death, were evaluated based on occurrences throughout the entire follow-up period. Efficacy was assessed in patients with clinically active UC and safety was analyzed in the overall population.
All statistical analyses were performed using R (version 4.2.1). The time to medication discontinuation was analyzed using Kaplan-Meier (KM) curves, constructed from the date of drug initiation to the date of discontinuation. Data were censored at the most recent follow-up. Survival estimates were compared using the log-rank test. p values <0.05 were considered significant.
Results
We identified 14, 36, and 13 elderly patients among 228, 215, and 159 UC patients treated with upadacitinib, filgotinib, and tofacitinib, respectively. Baseline characteristics showed no significant differences between the elderly and non-elderly groups for any of the three JAK inhibitors in terms of disease extent and duration, or follow-up time, while the proportion of elderly patients with prior advanced therapies, particularly vedolizumab, was higher compared to the non-elderly group in the filgotinib-treated patients. Regarding baseline disease activity, we have found no significant differences in clinical disease activity between the non-elderly and elderly groups for any of the JAK inhibitors. Although baseline endoscopic disease activity data were limited, no significant differences were observed between the two groups in the tofacitinib- or upadacitinib-treated patients. In contrast, the baseline Mayo endoscopic subscore was significantly higher in the elderly group compared to the non-elderly group in the filgotinib-treated patients. Albumin levels were significantly lower in the elderly group compared to the non-elderly group, consistent across all three JAK inhibitors (online suppl. Table 1; for all online suppl. material, see https://doi.org/10.1159/000546640).
In the upadacitinib-treated elderly group, 92.9% (13/14) started on 45 mg/day, and 78.6% (11/14) were switched to the maintenance dose (63.6%–30 mg/day and 36.4% to 15 mg/day). The median follow-up was 52 weeks (IQR 32–73), with a discontinuation rate of 29%. The KM curve showed no significant difference in time to discontinuation of upadacitinib between the elderly and non-elderly groups (p = 0.87) (online suppl. Fig. 1A, B). In the filgotinib-treated elderly group, 97.2% (35/36) started at 200 mg/day. The median follow-up was 48 weeks (IQR 13–64), with a discontinuation rate of 42%. The KM curve showed no significant difference in time to discontinuation of filgotinib between the elderly and non-elderly groups (p = 0.75) (online suppl. Fig. 1C, D). In the tofacitinib-treated elderly group, 100% (13/13) started on 20 mg/day, and 69.2% (9/13) were switched to the maintenance dose. The median follow-up was 175 weeks (IQR 11–288), with a discontinuation rate of 38%. The KM curve showed no significant difference in time to discontinuation of tofacitinib between the elderly and non-elderly groups (p = 0.18) (online suppl. Fig. 1E, F).
A comparative efficacy analysis based on a per-protocol analysis showed no significant differences in the rates of clinical remission, clinical response, or CS-free remission at any time point between the elderly and non-elderly groups for any of the three JAK inhibitors (Fig. 1, online suppl. Table 1). Since some data were missing, particularly at 26 and 58 weeks, we performed a sensitivity analysis based on an intention-to-treat approach to confirm the robustness of our findings. The analysis revealed no significant differences in these outcomes at any time point between the elderly and non-elderly groups across all three JAK inhibitors (online suppl. Figure 2). We also assessed endoscopic improvement at the most recent follow-up and found no significant differences between the elderly and non-elderly groups for any of the three JAK inhibitors. Despite higher baseline endoscopic disease activity in the elderly group among filgotinib-treated patients, the proportion of patients achieving endoscopic improvement at the most recent follow-up was comparable between the two groups (Fig. 1; online suppl. Table 1).
Fig. 1.
Bar graphs showing clinical remission, clinical response, and corticosteroid (CS)-free remission at 10 weeks, 26 weeks, and 58 weeks, as well as at the most recent follow-up, after initiation of upadacitinib (a), filgotinib (b), and tofacitinib (c). Endoscopic improvement is included at the most recent follow-up. Outcomes were stratified into patients starting a Janus kinase (JAK) inhibitor <65 years of age (non-elderly group) and patients starting a JAK inhibitor ≥65 years of age (elderly group). Data were analyzed by using a per-protocol approach. Fisher’s test showed no statistically significant differences in any of the outcomes between the non-elderly and elderly groups.
Safety analysis showed no significant differences between the elderly and non-elderly groups in terms of past history of HZ, shingles vaccination, or thrombosis for any of the three JAK inhibitors. The proportion of elderly patients with a past history of thrombosis was 7.1% (1/14) in the upadacitinib group, 5.6% (2/36) in the filgotinib group, and 0% (0/13) in the tofacitinib group (online suppl. Table 1). While the risks of colectomy and hospitalization were similar or lower in the elderly group compared to the non-elderly group, the risk of discontinuation due to intolerance was 2–3 times higher in the elderly group for each of the three drugs. Importantly, while the risk of HZ with filgotinib was less than 3% in both the elderly and non-elderly groups, the elderly group had a 3-fold higher risk of HZ compared to the non-elderly group in patients treated with upadacitinib (14% vs. 4.3%) or tofacitinib (23% vs. 6.8%). Among elderly patients who discontinued medication, 25% (1/4) and 20% (1/5) discontinued upadacitinib and tofacitinib, respectively, due to HZ, whereas only 6.7% (1/15) of the elderly patients discontinued filgotinib due to HZ. Conversely, the non-elderly group had a 3-fold higher risk of acne with upadacitinib compared to the elderly group (24% vs. 7.1%). Notably, no cases of thrombosis, MACE, malignancy, or death were reported in the elderly group with any of the three JAK inhibitors (Table 1).
Table 1.
Safety profile
| | Upadacitinib | Filgotinib | Tofacitinib | ||||||
|---|---|---|---|---|---|---|---|---|---|
| non-elderly, N = 2141 | elderly, N = 141 | p value2 | non-elderly, N = 1791 | elderly, N = 361 | p value2 | non-elderly, N = 1461 | elderly, N = 131 | p value2 | |
| Discontinuation | 55 (26%) | 4 (29%) | 0.764 | 73 (41%) | 15 (42%) | >0.999 | 92 (63%) | 5 (38%) | 0.135 |
| Reason for discontinuation | | | 0.076 | | | 0.130 | | | 0.455 |
| Ineffective | 33 (60%) | 1 (25%) | | 61 (85%) | 9 (64%) | | 70 (79%) | 4 (80%) | |
| Intolerance | 13 (24%) | 3 (75%) | | 9 (13%) | 4 (29%) | | 7 (7.9%) | 1 (20%) | |
| Others | 9 (16%) | 0 (0%) | | 2 (2.8%) | 1 (7.1%) | | 12 (13%) | 0 (0%) | |
| Colectomy | 10 (4.7%) | 0 (0%) | >0.999 | 4 (2.2%) | 1 (2.8%) | >0.999 | 9 (6.2%) | 0 (0%) | >0.999 |
| Hospitalization | 21 (9.8%) | 1 (7.1%) | >0.999 | 6 (3.4%) | 2 (5.6%) | 0.624 | 18 (12%) | 1 (7.7%) | >0.999 |
| HZ infection | 9 (4.3%) | 2 (14%) | 0.143 | 2 (1.1%) | 1 (2.8%) | 0.425 | 10 (6.8%) | 3 (23%) | 0.076 |
| Neutropenia3 | 7 (3.3%) | 0 (0%) | >0.999 | 5 (2.8%) | 1 (2.8%) | >0.999 | 1 (0.7%) | 0 (0%) | >0.999 |
| Acne | 52 (24%) | 1 (7.1%) | 0.198 | 8 (4.5%) | 0 (0%) | 0.357 | 9 (6.2%) | 1 (7.7%) | 0.585 |
| Thrombosis | 3 (1.4%) | 0 (0%) | >0.999 | 1 (0.6%) | 0 (0%) | >0.999 | 0 (0%) | 0 (0%) | |
| Major adverse cardiovascular events | 0 (0%) | 0 (0%) | | 0 (0%) | 0 (0%) | | 0 (0%) | 0 (0%) | |
| Malignancy | 0 (0%) | 0 (0%) | | 0 (0%) | 0 (0%) | | 2 (1.4%)4 | 0 (0%) | >0.999 |
| Death | 0 (0%) | 0 (0%) | | 0 (0%) | 0 (0%) | | 0 (0%) | 0 (0%) | |
1Frequency (%).
2Fisher’s exact test.
3Neutropenia was defined as a decrease in the neutrophil count below the lower limit of the normal range.
4Colitis-associated cancer and chronic myeloid leukemia.
Discussion
This comparative analysis demonstrated that clinical remission, clinical response, CS-free remission at each time point, and endoscopic improvement at the most recent follow-up were comparable between elderly and non-elderly UC patients treated with any of the JAK inhibitors. The safety analysis revealed a higher risk of HZ in the upadacitinib- or tofacitinib-treated elderly groups compared to the non-elderly groups, often leading to discontinuation of the drug, whereas both age groups had a low risk of HZ in the filgotinib-treated patients. These findings suggest that shingles vaccination should be recommended for patients over 65 years of age when prescribing JAK inhibitors, particularly upadacitinib or tofacitinib.
We observed no cases of thrombosis, MACE, malignancy, or death among elderly patients treated with any of the JAK inhibitors, supporting the potential safety of JAK inhibitors in this population. This may be partly explained by appropriate patient selection, as only a small proportion of elderly patients had a past history of thrombosis across the groups (7.1% for upadacitinib, 5.6% for filgotinib, and 0% for tofacitinib). However, given the limited number of elderly patients in our study, we acknowledge that these results should be interpreted with caution. It is important to consider findings from pivotal clinical trials assessing the safety of tofacitinib in high-risk populations. For instance, the ORAL Surveillance study compared the safety of tofacitinib versus tumor necrosis factor inhibitors in patients with active rheumatoid arthritis aged ≥50 years with at least one additional cardiovascular risk factor, despite methotrexate treatment. This study demonstrated higher risks of MACE and malignancy with tofacitinib compared to tumor necrosis factor inhibitors, and did not meet non-inferiority criteria [2]. These results underscore the need for careful consideration when prescribing JAK inhibitors to elderly patients with cardiovascular risks. Since such adverse events are dose-dependent, reduced induction doses of JAK inhibitors may offer a safe and effective option for elderly patients [3]. Further research is needed to determine optimal strategies for using JAK inhibitors in elderly patients with UC.
In our elderly cohort, filgotinib demonstrated a lower incidence of HZ and no reported cases of serious adverse events, which may support its consideration as a JAK inhibitor option for elderly patients. On the other hand, integrated safety data from the SELECTION and SELECTIONLTE studies also showed that the exposure-adjusted incidence rates (EAIRs) for serious infections, thromboembolic events, and MACE were consistently low across all treatment groups. However, the EAIR for HZ was numerically higher in the filgotinib 200 mg group compared with placebo. Notably, the incidence was higher in the elderly population (EAIR: 2.93) than in the non-elderly population (EAIR: 1.32) [9]. While these findings suggest that filgotinib is generally well tolerated and associated with no unexpected safety signals, further studies are warranted to determine whether filgotinib may be a preferable JAK inhibitor option for elderly patients.
We also found the non-elderly group treated with upadacitinib for UC was more likely to develop acne than the elderly group. Consistently, several studies assessing the safety of upadacitinib in patients with atopic dermatitis have shown that younger patients are at a higher risk of developing acne [10, 11]. A meta-analysis has shown a significant association of acne with depression and anxiety [12], which can lead to poor treatment adherence and suboptimal IBD control [13]. Thus, careful explanation and appropriate management of acne in younger patients are important.
To our knowledge, this is the first study to evaluate the efficacy and safety of the three JAK inhibitors in elderly patients. However, we acknowledge the limitations of this study, including its retrospective design, sample size, and incomplete data. For instance, while there appeared to be differences in the reasons for drug discontinuation between age groups – specifically, a higher rate of discontinuation due to lack of effectiveness in the non-elderly group and due to intolerance in the elderly group – further statistical analysis was limited by the relatively small number of elderly patients. Furthermore, some follow-up clinical data, particularly at 26 and 58 weeks, as well as endoscopic data were missing. For efficacy analysis at each time point, we conducted both per-protocol and intention-to-treat analyses. These analyses confirmed no significant differences in efficacy between the elderly and non-elderly groups for any of the three JAK inhibitors, suggesting that the influence of missing data on our findings appears minimal.
In conclusion, the elderly group had a higher risk of HZ with upadacitinib or tofacitinib but a lower risk of HZ with filgotinib. Meanwhile, the non-elderly group had a higher risk of acne with upadacitinib, suggesting that adverse event risks with JAK inhibitors should be considered by age. Further studies with larger sample sizes are needed to validate our findings.
Statement of Ethics
This study was approved by the Ethics Committees of all participating institutions: University of Tsukuba (R04-289), Institute of Science Tokyo (M2022-285-01), Kitasato University School of Medicine (E2023022), the Jikei University School of Medicine (35-126 (11752)), Iwate Medical University (C-IRB2023-007), Juntendo University School of Medicine (C23-0038-H02), Okayama University Hospital (2306-021), Tokyo Yamate Medical Center (K-029), Ofuna Chuo Hospital (No. 2024-002), Kurume University School of Medicine (22586), and Tokyo Women’s Medical University (C24-026). The Ethics Committees approved the use of an opt-out consent approach and waived the requirement for informed consent because this study retrospectively analyzed existing clinical data. All efforts were made to ensure the confidentiality of the data.
Conflict of Interest Statement
H.S., A.T., and T.S.: The Jikei University School of Medicine; M.K.: University of Tsukuba Hospital and Tokyo Women's Medical University; M.E. and S.Y.: Iwate Medical University; A.Y., K.N., and T.S.: Juntendo University School of Medicine; M.T. and K.T.: Kurume University School of Medicine; S.Y.: Kurume University School of Medicine; M.S.: Tokyo Yamate Medical Center; and T.F. have no relevant disclosures. S.A. has received honoraria (lecture fees) from AbbVie and EA Pharma. S.H. has received honoraria (lecture fees) from EA Pharma and AbbVie. K.Y. has received honoraria (lecture fees) from Takeda, Gilead, EA Pharma, and AbbVie. T.O. has received honoraria (lecture fees) from AbbVie and Takeda. M.S.: (The Jikei University School of Medicine) has received honoraria (lecture fees) from Mitsubishi-Tanabe, Takeda, Janssen, Gilead, EA Pharma, AbbVie, Mochida, Nobelpharma, Viatris, and Kissei; research grants from AbbVie, CMIC, and Shin Nippon Biomedical Laboratories; and scholarship grants from Zeria, Mochida, AbbVie, and Kissei. T.M. has received honoraria (lecture fees) from Mitsubishi-Tanabe, Takeda, Janssen, AbbVie, Gilead, Mochida, and EA Pharma and scholarship grants from Mitsubishi-Tanabe and Nippon Kayaku. R.O. has received research grants from Medical and Biological Laboratories and from EA Pharma and scholarship grants from Kyorin, Miyarisan, Zeria, and Mochida. K.T. (University of Tsukuba) has received honoraria (lecture fees) from Takeda, AbbVie, EA Pharma, and Janssen and research grants from Eli Lilly.
Funding Sources
The authors declare no funding was obtained or used for this study.
Author Contributions
Shintaro Akiyama designed the study, conducted the literature search, led the study, collected and interpreted the data, performed all the data analysis and created all the figures and tables, drafted the article, critically revised the manuscript, and approved the submitted version of the manuscript. Hiromichi Shimizu, Akiko Tamura, Kaoru Yokoyama, Toshiyuki Sakurai, Mariko Kobayashi, Makoto Eizuka, Shunichi Yanai, Kei Nomura, Tomoyoshi Shibuya, Masahiro Takahara, Sakiko Hiraoka, Minako Sako, Atsushi Yoshida, Kozo Tsuruta, Shinichiro Yoshioka, Miki Koroku, and Teppei Omori collected and interpreted the data, and approved the submitted version of the manuscript. Masayuki Saruta, Takayuki Matsumoto, Ryuichi Okamoto, and Kiichiro Tsuchiya read and approved the submitted version of the manuscript. Toshimitsu Fujii is the principal investigator, designed the study, conducted the literature search, led the study, collected and interpreted the data, critically revised the manuscript, and approved the submitted version of the manuscript.
Funding Statement
The authors declare no funding was obtained or used for this study.
Data Availability Statement
All data generated or analyzed during this study are included in this published article. Further inquiries can be directed to the corresponding author.
Supplementary Material.
Supplementary Material.
Supplementary Material.
References
- 1. Danpanichkul P, Duangsonk K, Ho AH, Laoveeravat P, Vuthithammee C, Dejvajara D, et al. Disproportionately increasing incidence of inflammatory bowel disease in female patients and the elderly: an update analysis from the global burden of disease study 2021. Am J Gastroenterol. 2025;120(3):675–80. [DOI] [PubMed] [Google Scholar]
- 2. Ytterberg SR, Bhatt DL, Mikuls TR, Koch GG, Fleischmann R, Rivas JL, et al. Cardiovascular and cancer risk with tofacitinib in rheumatoid arthritis. N Engl J Med. 2022;386(4):316–26. [DOI] [PubMed] [Google Scholar]
- 3. Viola A, Li Voti R, Bivacqua C, De Francesco C, Muscianisi M, Costantino G, et al. Mitigating the risk of tofacitinib-induced adverse events in the elderly population with ulcerative colitis. J Crohns Colitis. 2024;18(3):488–91. [DOI] [PubMed] [Google Scholar]
- 4. Khan N, Sundararajan R, Patel M, Trivedi C, Yang YX. Effectiveness of tofacitinib in patients with ulcerative colitis: a nationwide veterans administration cohort study. Am J Gastroenterol. 2024;119(8):1632–5. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5. Akiyama S, Shimizu H, Tamura A, Yokoyama K, Sakurai T, Kobayashi M, et al. Comparative efficacy and safety of three Janus kinase inhibitors in ulcerative colitis: a real-world multicentre study in Japan. Aliment Pharmacol Ther. 2025;61(3):524–37. [DOI] [PubMed] [Google Scholar]
- 6. Akiyama S, Yokoyama K, Yagi S, Shinzaki S, Tsuruta K, Yoshioka S, et al. Efficacy and safety of filgotinib for ulcerative colitis: a real-world multicenter retrospective study in Japan. Aliment Pharmacol Ther. 2024;59(11):1413–24. [DOI] [PubMed] [Google Scholar]
- 7. Walmsley RS, Ayres RC, Pounder RE, Allan RN. A simple clinical colitis activity index. Gut. 1998;43(1):29–32. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8. Schroeder KW, Tremaine WJ, Ilstrup DM. Coated oral 5-aminosalicylic acid therapy for mildly to moderately active ulcerative colitis. A randomized study. N Engl J Med. 1987;317(26):1625–9. [DOI] [PubMed] [Google Scholar]
- 9. Schreiber S, Rogler G, Watanabe M, Vermeire S, Maaser C, Danese S, et al. Integrated safety analysis of filgotinib for ulcerative colitis: results from SELECTION and SELECTIONLTE. Aliment Pharmacol Ther. 2023;58(9):874–87. [DOI] [PubMed] [Google Scholar]
- 10. Watanabe A, Kamata M, Okada Y, Tomura Y, Tada Y. Incidences of herpes zoster and acne by age in Japanese patients with atopic dermatitis during treatment with baricitinib or upadacitinib: a single-centre retrospective study. Exp Dermatol. 2025;34(4):e70100. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11. Hayashi N, Ikeda M, Liu J, Raymundo E, Liu Y, Sasaki T, et al. Acne among Japanese patients with atopic dermatitis receiving upadacitinib in the phase 3 rising up study. Dermatol Ther. 2023;13(8):1817–30. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12. Samuels DV, Rosenthal R, Lin R, Chaudhari S, Natsuaki MN. Acne vulgaris and risk of depression and anxiety: a meta-analytic review. J Am Acad Dermatol. 2020;83(2):532–41. [DOI] [PubMed] [Google Scholar]
- 13. Temido MJ, Honap S, Bursztejn AC, Portela F, Jairath V, Danese S, et al. Drug-Induced acne in inflammatory bowel disease: a practical guide for the gastroenterologist. Am J Gastroenterol. 2025;120(1):125–34. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
All data generated or analyzed during this study are included in this published article. Further inquiries can be directed to the corresponding author.