CYP2D6 Metabolizer Status Does Not Predict Response or Tolerability of Low-Dose Nortriptyline for Irritable Bowel Syndrome With Diarrhea Authors: Kyle Staller, Jenna Clukey, Elizabeth N Madva, Brian C Healy, Minyi Lee, Manish K Gala Published: 2026 Neuromodulators, including tricyclic antidepressants (TCAs), have long been used to treat irritable bowel syndrome with diarrhea (IBS-D),1,2 but clinical response to TCAs is not clearly related to dosage.3 Interindividual variations in adverse events and efficacy may be at least partially attributed to variability in drug plasma concentrations.4 Nortriptyline, a TCA commonly used in IBS-D, is primarily metabolized by CYP2D6. Genotype-inferred phenotype of CYP2D6 can classify patients into different metabolizer classes.5 Although genetic testing is widely available, it is unclear if these polymorphisms are relevant to the use of low-dose TCAs in IBS-D. We hypothesized that CYP2D6 metabolism may predict efficacy and side effects (SEs) of nortriptyline in patients with IBS-D. We prospectively enrolled patients aged 18–65 years meeting Rome IV criteria for IBS-D from an outpatient tertiary gastroenterology clinic initiating treatment with nortriptyline as part of routine clinical care in a standardized step-up approach: starting at 10 mg at bedtime for 2 weeks, followed by an increase to 20 or 25 mg if no SEs and no improvement. If, after 2 additional weeks, there were no SEs or improvement, patients increased their dose to 50 mg (maximum dose) (Supplementary Methods). Patients were excluded if using other P450-metabolized antidepressants including selective serotonin reuptake inhibitors and selective serotonin-norepinephrine reuptake inhibitors. Patients completed baseline and biweekly assessments of symptom severity (IBS Symptom Severity Score [IBS-SSS]),6 and a standardized SEs screener for TCAs7 for 12 weeks. At enrollment, patients submitted a saliva sample for Blended Genome/Exome sequencing for CYP2D6 genotype at the Broad Institute (Cambridge, MA). Genotypes were translated to metabolizer phenotypes (ultrarapid, normal, intermediate, and poor metabolizer; genotypes for which phenotype could not be definitively assigned were classified as indeterminate) using standard classifications.5 The primary endpoint was time to a clinically significant improvement in IBS symptoms (≥50-point decrease in IBS-SSS). Kaplan-Meier curves with log-rank tests were used to determine differences in time to significant improvement by phenotype, while change in SEs was compared descriptively. We enrolled 45 patients with IBS-D (mean age 36.4, 82% female) starting nortriptyline who provided saliva samples and completed the 12-week study tracking period. The mean IBS-SSS score at baseline was 281.4 ± 108.3 (moderate severity). Seven subjects discontinued nortriptyline due to SEs, with a median time to discontinuation of 35 days (IQR: 27–44). Subjects discontinued at doses of 10 mg (n = 3) and 25 mg (n = 4). CYP2D6 genotype phenotypes included normal metabolizers (n = 32) where the recommendation is to initiate therapy with the recommended starting dose, intermediate metabolizers (n = 3) where the guidance is to consider a 25% reduction of the recommended starting dose, and indeterminate metabolizers (n = 10), where no recommendation is made. There were no significant differences in the genotype phenotype of those that discontinued (normal phenotype n = 4, 12.5%; intermediate phenotype n = 1, 33%, indeterminate n = 2, 20%) (P = .47). The majority (80%, n = 36) of study patients achieved a 50-point decrease in IBS-SSS scores, the threshold for meaningful clinical improvement, with the most doing so at the 10 mg dose (n = 27, 75.0%) and fewer at higher doses (20 mg [n = 3], 25 mg [n = 5], and 50 mg [n = 1]). Median time to clinically significant improvement was within 14 days. Time to clinically significant improvement did not differ by CYP2D6 genotype phenotype (Figure 1), but females and those with moderate IBS severity were more likely to demonstrate improvement (Supplementary Table 1). Median time to clinically significant improvement for the normal CYP2D6 phenotype was 28 days, 42 days for the intermediate phenotype, and 14 days for the indeterminate phenotype. Kaplan-Meier curve of symptom improvement by CYP2D6 phenotype. All but 2 patients reported symptoms before starting treatment that could be subsequently classified as SEs (n = 43, 95.6%). The top 3 most common SEs that were new or worsened at 10 mg were dry mouth (n = 21, 58.3%), trouble sleeping (n = 14, 38.9%), and fatigue (n = 13, 36.1%). Percent change in SEs from baseline was not significantly different by CYP2D6 phenotype (P = .085) (Figure 2). Those that discontinued did not have significantly more new or worsening SEs at 10 mg than those that did not discontinue at 10 mg (3.1 vs 2.5 [−3.2–1.9], P = .56). The most common SE at 20, 25, and 50 mg doses of nortriptyline was dry mouth (75.0% n = 3, 63.3% n = 19, 90%, n = 9, respectively). Some participants also reported decreases in reported symptom severity at 10 mg from those symptoms reported at baseline (n = 36, 80%). The top 3 symptoms that improved from baseline were fatigue (n = 17, 47.2%), headaches (n = 15, 33.3%), and trouble sleeping (n = 14, 31.1%). Percent change in side effects at the 10 mg nortriptyline dose compared with baseline, by CYP2D6 phenotype. In a prospective, observational study of patients with IBS-D starting low-dose nortriptyline using a standardized step-up protocol, we found that a majority of patients had a successful clinical response independent of CYP2D6 phenotype, though genotype diversity was limited. Similarly, commonly reported TCA SEs and likelihood of medication discontinuation did not differ based on phenotype. Because of their ability to neuromodulate the activity of the peripheral and central nervous systems, TCAs have long played a role in the pharmacologic treatment of IBS-D after conservative treatments have failed. However, enthusiasm for their use is at least partly limited by prescriber concern about SEs and studies from the psychiatry literature showing high rates of discontinuation.1 These studies used much higher doses than are used in IBS-D, for which TCA doses are typically less than 50 mg daily. A randomized controlled trial of nortriptyline in functional dyspepsia selected only normal metabolizers by CYP2D6 phenotype found nortriptyline not more effective than placebo, with no significant differences in adverse events.8 In our study, many SEs that would normally be attributed to TCAs were present at baseline, and some of these improved with nortriptyline treatment—a pattern seen in other studies as well.9 Our study has several strengths, most notably our prospective design with capture of time to improvement and incident SEs in real time. We also acknowledge some limitations, specifically the lack of a placebo arm and the relative lack of genotype diversity, which may be particularly relevant for ultrarapid TCA metabolizers, of which we had none; thus, our findings may not be generalizable to less common metabolizer phenotypes. However, an aggregated pool of individuals in the United States found that 91.4% of patients had normal CYP2D6 metabolizer status.10 In summary, the TCA metabolizer genotype does not appear to influence the efficacy or SEs of IBS-D treatment with low-dose nortriptyline. Future studies with more diverse metabolizer genotypes may be helpful in determining whether this guidance can be applied more broadly. Acknowledgments Authors’ Contributions Kyle Staller and Manish K. Gala led the study design. Kyle Staller and Jenna Clukey analyzed the data. Kyle Staller drafted the manuscript and supervised the study. All authors contributed to the interpretation of the results and revision of the manuscript critically for important intellectual content. All authors had full access to all data (including statistical reports and tables) in the study and can take responsibility for the integrity of the data and the accuracy of the data analysis. Footnotes Conflicts of Interest: This author discloses the following: Kyle Staller has received research support from Ardelyx and Ironwood and has served as a consultant to AbbVie, Ardelyx, Atmo, Ferring, Gemelli, Laborie, Mindset Health, Salix, and Takeda. The remaining authors disclose no conflicts. Funding:NIDDKK23DK120945 (Kyle Staller) and NIDDKR01DK137971 (Manish K. Gala). Ethical Statement: The study protocol was approved by the institutional review board of Massachusetts General Hospital (Protocol #2016P002242). Data Transparency Statement: The data of this study are available upon reasonable request. Reporting Guidelines: Reporting Guidelines were not applicable for this article type. Material associated with this article can be found, in the online version, at https://doi.org/10.1016/j.gastha.2026.101111. Supplementary Materials Supplementary Table 1 Supplementary Materials Extended PDF References Associated Data Supplementary Materials Supplementary Table 1 Supplementary Materials Extended PDF