Efficacy and safety of naldemedine for opioid-induced constipation in older patients with cancer: a retrospective study

Background Opioids are pain relievers that are often associated with opioid-induced constipation (OIC) that worsens with age. We performed a multicenter, retrospective analysis on the efficacy and safety of naldemedine, an opioid receptor antagonist, in treating OIC in patients with cancer (age >75 years). Methods The electronic medical records of cancer patients who received naldemedine at 10 Japanese institutions between 7 June 2017 and August 31, 2019, were retrieved. Patients aged ≥75 years who were treated with naldemedine for the first time and hospitalized for at least 7 days before and after initiating naldemedine therapy were included in this analysis. Results Sixty patients were observed for at least 7 days before and after starting naldemedine. The response rate was 68.3%, and the frequency of bowel movements increased significantly after naldemedine administration in the overall population (P < 0.0001) and among those who defecated <3 times/week before naldemedine administration (P < 0.0001). Diarrhea was the most frequent adverse event in all grades, observed in 45% of patients, of which 92.6% were Grade 1 or 2. Grade 4 or higher adverse events, including death, were not observed. Conclusion Naldemedine exhibits significant efficacy and safety in OIC treatment in older patients with cancer.


Background
Opioids are morphine-like substances used for chronic pain treatment, active-phase cancer treatment, and palliative care; however, opioid-induced constipation (OIC) remains a common side effect, occurring in more than half the opioid recipients if not prevented with laxatives or other medications [1,2].Unlike other adverse events such as nausea and vomiting, OIC does not develop tolerance with continued opioid administration [3].Moreover, the risk of anorexia, abdominal pain, delirium, vomiting, and nausea increases with prolonged constipation [4], thus reducing the quality of life and affecting its use as an analgesic or rescue medication, with significant negative consequences on pain control [5].Therefore, appropriate management of OIC is crucial to maintain adequate quality of life in patients with cancer under opioid therapy.OIC is also classified as functional constipation and is defined as changes in bowel habits and defecation patterns from the baseline before opioid therapy initiation to after opioid therapy initiation [6].A study based on Rome IV, a set of international diagnostic criteria for OIC, reported a 56% incidence of OIC in Japan for various malignancies, irrespective of age, performance status (PS), and type of opioid analgesic [7].
Opioid analgesics act on the central μ-opioid receptor; activation of μ-opioid receptors in the intestinal tract inhibits normal bowel movement [8].Moreover, opioidinduced bowel dysfunction has been reported to occur quickly after opioid administration [9].Naldemedine is a peripheral µ-opioid receptor antagonist (PAMORA) that does not cross the blood-brain barrier and is expected to improve OIC without interfering with its analgesic effects [10].In phase III randomized trials of naldemedine, the number of spontaneous bowel movements (SBMs) without straining or residual sensation was significantly increased compared with the placebo, and SBM was observed in many patients within 12 h after initiation of the drug [11,12].Discontinuation of naldemedine administration due to toxicities was reported in 9.3% of the patients, with the most frequent adverse events being diarrhea (19.6%), malaise (4.1%), vomiting (3.1%), and decreased appetite (3.1%) [11].
Patients aged 70 years and older represent 42% of all cancer patients [13].Even with no clear medical or biological evidence, in many countries, old age is defined as a chronological age of 65 or more.However, several academic societies in Japan have reported proposals to define old age as 75 years or older [14].Constipation is common in the older population, even in the absence of opioid use, owing to changes in lifestyle, exercise habits, and comorbidities [15,16].Moreover, older individuals have decreased colonic transport capacity (colonic transit time) [17].Therefore, great care should be employed while treating OIC in older individuals.Various factors influence constipation in old age.Most prospective trials are carefully conducted in selected patients with good general condition and organ function; in the selected older patients, the predictors of efficacy or adverse events were not examined in detail.Published randomized phase III trials of naldemedine have included patients with favorable ECOG-PS of 0-2 and age <75 years; therefore, relevant clinical data on patients older than 75 years are unavailable.Older patients are treated based on the results of these prospective clinical studies, and the tolerability of naldemedine in such patients remains unclear.We have previously reported naldemedine in clinical practice [18,19], albeit with insufficient findings on its efficacy and toxicity in older patients with cancer.Herein, we retrospectively evaluated the efficacy and safety of naldemedine when treating OIC in older patients with cancer aged ≥75 years who also received opioid therapy.

Patients
Clinical data of older patients with cancer treated with naldemedine at 10 Japanese institutions between June 7, 2017 and August 31, 2019, were retrospectively examined from electronic medical record data.Data of patients who received naldemedine for the first time during hospitalization were extracted from pharmacy databases.The inclusion criteria for the study are (1) hospitalization for more than 7 days before and after initiating naldemedine treatment with evaluated number of bowel movements; (2) concomitant naldemedine use with opioids; and (3) patients with cancer aged ≥75 years.We evaluated patient characteristics and data on treatment response to naldemedine and its adverse events.This study's sample partially overlaps that of a previous study, as the sub-analysis is limited to older patients [19].

Treatment
Patients received 0.2 mg of naldemedine orally, once daily, concomitantly with opioids.The treatment was continued until unacceptable toxicity developed, the patient withdrew consent for treatment, or the attending physician determined that treatment needed to be discontinued for any reason.The initiation, termination, or discontinuation of the treatment was determined at the discretion of each attending physician.

Assessment of treatment efficacy and adverse events
We assessed the frequency of bowel movements (times/ week) for 7 days before and after naldemedine administration.Defecation not induced by rescue laxatives was evaluated as bowel movements.Patients who had three or more bowel movements/week, and an increase of one or more bowel movements/week from baseline for 7 days after naldemedine administration were defined as responders.The number of bowel movements during the week before initiating naldemedine treatment was considered the baseline.Adverse events were evaluated with the Common Terminology Criteria for Adverse Events version 5. BMI (weight [kg]/height [m 2 ]) was assessed before the start of naldemedine administration.We evaluated the potential correlation between BMI and naldemedine effectiveness based on a BMI cutoff value of 22.0 kg/m 2 , that is, the ideal BMI for the Japanese population [20] (high and low BMI: ≥22.0 and <22.0 kg/m 2 , respectively).

Statistical analyses
Fisher's exact test was used to test associations between categorical variables.The Wilcoxon signed-rank test was used to evaluate normality and equal variances and test the correspondence between the two groups.Factors predicting efficacy were identified and evaluated using multivariate ordered logistic regression analysis and expressed as odds ratios (ORs) with 95% confidence intervals (CIs).Differences were considered significant at a two-tailed P-value of <0.05.All statistical analyses were performed using JMP version 11.0 for Windows (SAS Institute, Cary, North Carolina, USA).

Patient backgrounds
In total, 387 patients who first received oral naldemedine therapy during hospitalization were selected from the database.Of these, 230 patients whose number of bowel movements could not be confirmed for at least 7 days before or after initiating naldemedine and eight patients for other reasons (non-cancer patients, prescribed but not taken, not under concomitant opioid use) were excluded from the analysis.Eighty-nine patients aged <75 years were also excluded from the study but were considered the comparator group for efficacy and safety.Finally, 60 patients were enrolled in the current study (Figure S1, supplemental digital content 1, http://links.lww.com/EJGH/B4).Forty-six of the analyzed patients died by the data-cutoff date.The clinical backgrounds of the enrolled patients are displayed in Table 1.The median age was 80 years (range, 75-96 years), with 30 (50.0%) patients aged 75-79 years and 30 (50.0%) aged ≥80 years.The males and females in the study were 38 (63.3%) and 22 (36.7%),respectively, with males being predominant.According to the ECOG-PS criteria, 25 (41.7%)patients had a PS of 0-1, 35 (58.3%) had a PS of 3-4, and most patients showed poor PS.The most common primary tumors in the patient population included lung, hepatobiliary pancreatic, and gastrointestinal cancers, with 13 patients (21.7%) exhibiting each cancer.
A breakdown of the opioid and laxative dosing details is displayed in Table 2.The median opioid dose in oral morphine equivalent was 27 mg/day (7.5-600 mg).The most commonly used opioid was oxycodone in 37 (61.7%)patients, followed by fentanyl in 13 (21.7%).

Treatment effectiveness
There were 41 responders and 19 non-responders, represented in Fig. 1 as a pie chart.The responder rate was 68.3% (95% CI: 56.5-80.1%).Table 3 shows the comparative patient backgrounds according to treatment response (responder/non-responder). No statistical difference between responders and non-responders was observed for any clinical factors analyzed.Patients under 75 years of age were also evaluated; they were considered as the control group which consists of 57 responders and 32 non-responders as represented in a pie chart in Figure S2, supplemental digital content 1, http://links.lww.com/EJGH/B4.The responder rate was 64.0% (95% CI: 54.0-74.0%).No significant differences were found in responder rates between the two age groups (P = 0.60).
Next, we examined changes in the number of bowel movements before and after starting naldemedine therapy in a group of patients that included the following overall population: all patients and only patients who had less than three bowel movements per week prior to naldemedine administration and were diagnosed with constipation prior to initiating naldemedine administration (Fig. 2).According to the overall population analysis (N = 60), the median number of bowel movements in the week before and after naldemedine administration was three (range, 0-14) and seven (range, 0-21), respectively, suggesting a significant increase in the number of bowel movements after naldemedine treatment (P < 0.0001; Fig. 2a).We also compared the number of bowel movements a week before and a week after naldemedine administration in patients who had less than three bowel movements per week before naldemedine administration (N = 22).The median number of bowel movements during a week before and a week after naldemedine administration was one (range, 0-2) and four (range, 0-11), respectively; the number of bowel movements increased significantly after naldemedine treatment (P < 0.0001; Fig. 2b).

Safety
Table 4 lists the toxicities associated with naldemedine treatment.Of the observed toxicities, diarrhea was the most frequent adverse event of any grade, observed in 27 patients (45.0%), of whom 25 (92.6%) were grade 1 or 2. No patients showed adverse events of grade 4 or higher, including death.Adverse events in the patient group under 75 years of age also included diarrhea, occurring in 30 patients (33.7%), of whom 27 (90.0%)were grade 1 or 2 (Table S1, supplemental digital content 2, http://links.lww.com/EJGH/B5).

Clinical factors influencing treatment efficacy
Multivariate logistic regression analysis of the correlation between naldemedine efficacy and various clinical factors is shown in Table 5.The clinical factors analyzed, including sex, age, PS, BMI, and concomitant use of laxatives before naldemedine treatment, were not statistically significant for naldemedine efficacy.

Discussion
In this study, we assessed the number of bowel movements and adverse events in older cancer patients treated with opioids, hospitalized for at least 7 days before and after starting naldemedine, and whose frequency of bowel movements could be evaluated.Moreover, changes in the frequency of bowel movements before and after naldemedine treatment were assessed to analyze the effect of naldemedine and influential clinical factors.To the best of our knowledge, previous studies have not evaluated the efficacy and safety of naldemedine in older patients with cancer.Herein, 68.3% of the patients responded to naldemedine treatment despite their old age.The responder rate in our cohort was comparable to that in the COMPOSE-4 study (71%) and to that of naloxegol, a PAMORA similar to naldemedine (73%) [11,21].The responder rate in our aged cohort is also comparable to that for the group aged less than 75 years (64%) analyzed during the same time in the current study.Because one of the diagnostic criteria for Rome IV is 'defecation frequency of less than three times per week', we focused on patients with less than three bowel movements per week and evaluated changes in the frequency of bowel movements.A statistically significant increase in the frequency of bowel movements was observed after naldemedine administration in patients who were constipated with less than three bowel movements per week, suggesting the effectiveness of naldemedine for older patients with cancer.Multivariate logistic regression analysis did not reveal any clinical factors (sex, age, PS, BMI, or concomitant use of laxatives before naldemedine treatment) influencing the effectiveness of naldemedine in older patients with cancer.Our findings are consistent with previous studies that evaluated clinical factors affecting naldemedine efficacy in younger OIC patients [22,23], indicating that the desired efficacy can be obtained irrespective of age.
The current study included patients with cancer aged ≥75 years, whereas the COMPOSE-4 and COMPOSE-5 phase III trials of naldemedine were conducted in patients with cancer and OIC aged ≥20 years [11].Although the median age of the patients included in these trials was approximately 64 years, data on the proportion of patients older than 75 years in these studies were unavailable.Therefore, the efficacy and safety of naldemedine in older patients aged ≥75 years have not been evaluated in prospective clinical trials.Older patients   are reportedly underrepresented in clinical trials, thus setting inaccurate standards for the efficacy and safety of cancer treatments [24,25].Only 24% of the participants in clinical trials registered with the US Food and Drug Administration are aged 70 years or older [24,25], and less than 10% of the patients in this age group get involved in National Cancer Institute-sponsored clinical oncology trials [26][27][28].Even when older patients are enrolled in cancer trials, they typically have fewer functional impairments or comorbid conditions [29] than the average older patient treated in clinical settings [26][27][28].Therefore, most risks and benefits of oncological care are based on the results of clinical trials conducted on younger, healthier patients [30], leading to systematic differences in treatment and disparities in health outcomes between older and younger patients with cancer [31][32][33].
Currently, three PAMORAs are regularly used to treat OIC, namely, methylnaltrexone, naloxegol, and naldemedine [34].However, no comparative studies of these drugs have been conducted, the superiority of one drug over the other remains unexplored, and not all of them are clinically used in many countries.Moreover, the criteria for evaluating the efficacy of a drug and its approval varies between the drugs.Therefore, the PAMORA best suited for older patients with cancer is not clear, especially in terms of efficacy and safety.Previous reports on naldemedine demonstrate that the baseline patient backgrounds affecting the drug efficacy in patients with OIC are poorly understood [22,23].Consistent with previous findings, the sex, age, PS, BMI, and concomitant use of laxatives before naldemedine treatment did not significantly affect the efficacy of naldemedine in older patients with cancer, indicating a promising clinical efficacy in older patients, regardless of baseline patient background.
From the perspective of safety, diarrhea and abdominal pain were the most common adverse events reported in prospective clinical trials of OIC in patients with cancer, including in young patients, with incidence rates of 19.6-39.7% and 1.7%, respectively [11,35].In the current study, the occurrence of diarrhea and abdominal pain in older patients was found to be 45.0% and 1.6%, respectively, comparable to those in a prospective phase III study involving all age groups, including young patients.Furthermore, the toxicity, including diarrhea, was comparable to that in our contemporaneous young group.Even though this cohort included patients aged 75 years and older, serious adverse events occurred in only two patients (3.3%, grade 3 diarrhea), suggesting that naldemedine can be feasibly administered to older patients with cancer in clinical practice.Therefore, OIC management using naldemedine in older patients with cancer does not require special caution, as no unique adverse events were observed for toxicity.
This study has several limitations.First, because of the retrospective design of this study, defecation-related assessments such as the bowel function index [36], Bristol stool form scale [37], and defecation diary were unavailable, which may have compromised the validity of the data.Therefore, further validation in clinical practice is needed.Second, the decision to start, stop, or skip naldemedine administration was left to the discretion of each attending physician.Furthermore, since the study was conducted on cancer patients, they received various cancer-related treatments, including surgery, radiation therapy, and chemotherapy; various types of opioids, and other concomitant medications, such as laxatives, were not standardized  enough for a retrospective study.Nevertheless, even though the study findings must be interpreted in light of the abovementioned limitations, we believe that the results closely simulate actual clinical practice in older patients with cancer, and are worth reporting.Third, naldemedine is administered to many outpatients in clinical practice; however, this study was limited to inpatients whose frequency of bowel movements could be assessed by the medical staff.Nonetheless, the inpatients' frequency of bowel movements is objectively evaluated and recorded by various healthcare professionals such as physicians, nurses, and pharmacists and is considered much more reliable than that reported by outpatients.

Conclusion
In this study, the clinical efficacy and feasibility of naldemedine treatment in older patients with cancer were similar to those of a prospective phase III study that included younger patients in the analysis.The favorable efficacy of naldemedine for treating OIC in older cancer patients and its safety with respect to adverse events make it a viable treatment option.

Fig. 2
Fig. 2 Comparison of the frequency of bowel movements.(a) Comparison of the frequency of bowel movements during the week before and after naldemedine administration in all patients (N = 60).(b) Comparison of the frequency of bowel movements during the week before and after naldemedine administration, limited to patients who had less than three bowel movements per week before naldemedine administration (N = 22).

Table 1 .
Patient characteristics a Within 3 weeks before starting naldemedine administration.

Table 2 .
Administration of opioids and laxative agents a Oral morphine equivalent of regular opioids.b Total number of patients.

Table 3 .
Patient characteristics according to response

Table 4 .
Adverse events during naldemedine administration a Graded using the Common Terminology Criteria for Adverse Events version 5.0.

Table 5 .
Multivariate logistic regression analysis of the factors specific to responders among patients receiving naldemedine