Efficacy of stepwise extended letrozole treatment in patients with resistant polycystic ovary syndrome undergoing intrauterine insemination: A 7-year retrospective cohort study
Article information
Abstract
Objective
Letrozole (LE) is the first-line therapy for ovulation induction in polycystic ovary syndrome (PCOS). However, in women who are resistant to standard LE therapy, alternative ovulation induction regimens require further exploration. This study evaluated the efficacy of a novel protocol for stepwise extended LE treatment in patients with anovulatory PCOS undergoing intrauterine insemination who were resistant to the standard dose of 5 mg LE.
Methods
The study included 319 women diagnosed with PCOS. LE treatment was administered in a stepwise manner, starting with 5 mg for 5 days (standard group), followed by an extension for another 2 to 5 days (total 7–10 days; extended group; LE 5 mg), and subsequently escalating the dose to 7.5 mg for 2–5 days (total 12–15 days; non-responder extended group; LE 7.5 mg) within a single cycle. Clinical outcomes included ovulation rate and clinical pregnancy rate (CPR).
Results
Overall, 219 patients responded to standard LE with follicular development, 83 responded to extended LE 5 mg therapy, and 11 responded to extended LE 7.5 mg. CPRs at 10 weeks of gestation were 18.72% in the standard group, 14.45% in the extended LE 5 mg group, and 11.76% in the non-responder extended LE 7.5 mg group, with no statistically significant differences among them. Multiple pregnancies were observed only in the standard group.
Conclusion
The stepwise extended LE regimen appears to be a feasible option for LE-resistant PCOS; however, its efficacy must be further explored and validated in prospective randomized studies.
Introduction
Anovulation is the most prevalent cause of subfertility, contributing to approximately 30% of female subfertility, with polycystic ovary syndrome (PCOS) accounting for 90% of these cases. PCOS is a complex endocrinological disorder with prevalence rates ranging from 6% to 21% in the global population, depending on phenotype and ethnicity [1]. The prevalence of PCOS is even higher among subfertile women, at approximately 20%–25% [2].
Ovulation induction (OI) is the first-line therapy for achieving pregnancy in a significant proportion of anovulatory women with PCOS. Clomiphene citrate (CC) has been used as a first-line medication for OI at a dosage of 50 to 150 mg for up to six cycles [3,4]. However, 20% to 40% of patients develop CC resistance, limiting its use [5]. Furthermore, the longer half-life of CC and its dual agonist-antagonist action on estrogen receptors can lead to reduced conception rates [6].
In 2018, new evidence-based international consensus guidelines for women with PCOS highlighted the use of letrozole (LE) as first-line treatment for OI to improve pregnancy rates while reducing adverse effects [7,8]. Exogenous gonadotropins are recommended as a second-line option for individuals who do not respond to initial treatment. However, gonadotropin use in LE-resistant patients is constrained by factors such as limited availability, the requirement for skilled clinicians, economic considerations, and the risk of developing multiple follicles [9,10].
LE, a third-generation aromatase inhibitor, effectively blocks the production of estrogen by reversibly binding to the enzyme aromatase, thereby reducing the conversion of androgens to estrogens [11]. Compared to CC, LE has a shorter half-life and does not affect estrogen receptors [12]. The traditional LE treatment protocol involves administering 2.5 mg daily for 5 days. If follicular development is not observed, the dosage can be escalated to 5 or 7.5 mg in subsequent cycles [13,14]. Nonetheless, a substantial number of patients with PCOS demonstrate resistance to conventional LE treatment protocols [15].
Recent research indicates that extending the duration of LE treatment in patients with PCOS who have not ovulated with the standard 5-day LE regimen can promote follicular development and support ovulation. This novel approach seeks to overcome the challenges posed by patients with resistant PCOS, potentially improving their chances of a successful pregnancy [16]. To further optimize extended LE treatment, the present study developed a novel protocol involving stepwise intracycle extended and dose-escalated LE to improve follicular development, ovulation, and clinical pregnancy rates in patients with highly resistant PCOS undergoing intrauterine insemination (IUI) in a tertiary care setting.
Methods
1. Study design and characteristics
This retrospective cohort study was conducted at the Krishna Institute of Medical Sciences (KIMS) Fertility Center, KIMS, Hyderabad, India, between 2016 and 2023. The inclusion criteria included women aged 21 to 40 years who had been diagnosed with anovulatory PCOS based on the Rotterdam criteria. The exclusion criteria comprised women with chronic medical conditions such as chronic heart disease, organ dysfunction, or neurological disorders, as well as those undergoing other pharmacological OI treatments.
2. Ethical approval
The study was approved by the institutional ethics review board of the Krishna Institute of Medical Sciences Hospital (KIMS/EC/BMHR/2023/69-08), dated October 3, 2023.
3. Treatment protocol
Detailed infertility evaluations were conducted for both male and female partners. For the women, height, weight, and other anthropometric measures were recorded, and endocrinological parameters were measured, including basal serum levels of follicle-stimulating hormone (FSH), luteinizing hormone (LH), estradiol, prolactin (PRL), thyroid-stimulating hormone (TSH), and total testosterone (TT). As per the institution’s protocol, a glucose tolerance test (GTT) was performed for all patients diagnosed with PCOS. The male partners underwent semen analysis and blood tests in accordance with protocol.
The first treatment cycle was initiated with the administration of 5 mg/day of LE beginning on the second day of the menstrual cycle (either a spontaneous period or a progesterone-induced withdrawal bleed) for a total of 5 days (‘standard protocol’) to assess ovarian response. When dominant follicles (>10 mm) were observed, gonadotropin doses ranging from 37.5 to 75 IU were administered to further promote multifollicular growth. Serial scans were performed meticulously to document the number and size of follicles, as well as endometrial thickness. Ovulation was triggered once the leading follicle reached 18 mm in size using either recombinant human chorionic gonadotropin (hCG; Ovitrelle, Merck KGaA) or a gonadotropin-releasing hormone agonist (Decapeptyl; Debiopharm), followed by IUI.
Patients who did not respond to the standard LE protocol were given LE 5 mg/day for a further 2 to 5 days without intervening withdrawal bleeding (extended LE group; total 7 to 10 days). In patients who still showed no response, the dosage was increased to 7.5 mg and administered for an additional 2 to 5 days (total 12 to 15 days) (non-responder extended LE group; 7.5 mg). Once a dominant follicle was noted in patients resistant to standard LE, further treatment followed the same protocol as in the standard group.
The occurrence of ovulation was confirmed and documented by ultrasound. Semen was prepared using a two-layer density gradient separation technique. The processed sample was then suspended in 0.5 mL of media to facilitate the insemination procedure.
4. Outcome variables
The primary outcomes were ovulation rate and clinical pregnancy rate at 10 weeks of gestation. The secondary outcomes were the number of dominant follicles >14 mm on the day of trigger, endometrial thickness, and multiple pregnancy.
5. Statistical analysis
Statistical analysis was conducted using SPSS ver. 21 (IBM Corp.). Categorical variables were described using frequencies and percentages. The Student t-test was used to assess continuous variables; these were presented as mean±standard deviation for normally distributed data and as median (interquartile range) for non-normally distributed data. The chi-square test and the Fisher exact test were applied to evaluate relationships between groups and variables. Binary logistic regression was employed to identify predictors and their impact on a binary outcome. Receiver operating characteristic curve analysis was performed to assess how accurately the model predicted a binary outcome. The area under the curve was calculated to assess overall test accuracy. A p-values of less than 0.05 was considered to indicate statistical significance, while p-values greater than 0.05 indicated the lack of significance.
Results
1. Stepwise treatment protocol characteristics
A total of 319 women diagnosed with PCOS who fulfilled the inclusion and exclusion criteria were included in the study. Figure 1 outlines the specifics of the LE protocol framework implemented. Initially, all patients received LE at a dosage of 5 mg per day for 5 days. Of these, 219 patients who responded with the development of a dominant follicle were categorized as the standard protocol group, while 100 patients did not respond. Subsequently, the non-responders (n=100) received extended LE therapy at 5 mg/day for an additional 2 to 5 days. Of these 100 patients, 83 responded to the extended therapy. The 17 patients who did not respond to the extended therapy of LE 5 mg were immediately given an increased dose of 7.5 mg daily for a further 2 to 5 days. Of these 17 patients, 11 responded to extended LE 7.5 mg therapy, whereas only six patients did not respond to the escalated intracycle LE dosage.
Schematic illustration of the stepwise letrozole protocol framework and clinical outcomes. Initially, patients (n=319) received the standard letrozole dose of 5 mg for 5 days. Responders (n=219) developed dominant follicles >10 mm, while non-responders (n=100) underwent extended protocols. Non-responders received extended letrozole 5 mg for an additional 2 to 5 days (n=83), resulting in ovulation (n=79), clinical pregnancy at 10 weeks (n=12), and live birth (n=11). Patients who did not respond to extended letrozole (n=17) received an escalated letrozole dose of 7.5 mg for an additional 2 to 5 days, resulting in ovulation (n=11), clinical pregnancy (n=2), and live birth (n=2). PCOS, polycystic ovary syndrome.
2. Study characteristics
The mean age of participants in the standard protocol group was 28.3±3.8 years, while that in the extended protocol group was 28.4±3.3 years. The extended group had a higher body mass index (BMI), at 27.54±4.85 kg/m2, compared to the standard group (p=0.007). The duration of infertility was slightly higher in the extended group (2.06±1.39 years) than in the standard LE group (1.84±1.41 years), but no statistically significant difference was observed. Primary infertility was noted in 76.2% of the standard group and 74% of the extended group, while secondary infertility was observed in 23.7% and 26% of patients in the standard and extended groups, respectively. The length of the menstrual cycle differed significantly between groups (p<0.05). The history of prior OI cycles performed at other clinics was recorded, with the number of cycles ranging from 0 to more than 10 (p<0.05) (Table 1).
Study characteristics of the standard protocol and extended LE groups
3. Hormonal profile and semen analysis
The extended protocol group exhibited a significantly higher mean antral follicle count (AFC; 21.7±3.7) and anti-Müllerian hormone (AMH) level (7.9±4.89 ng/mL) compared to the standard protocol group, with statistical significance. Higher PRL concentrations (13.83±10.47 IU/L) were observed in the standard group (p=0.015). Hormonal and biochemical profiles, including FSH, LH, TSH, TT, and GTT, did not differ significantly between the two groups (p>0.05). Semen parameters, such as sperm volume, sperm count, and total motile count, were also comparable between groups (p>0.05) (Table 2).
Hormonal profile and semen analysis of the study participants
4. Treatment outcomes
The mean number of dominant follicles (>14 to ≤18 mm) on the day of trigger was 2.0±1.09 in the standard group, 1.75±1.14 in the extended LE 5 mg responder group, and 2.0±0.70 in the extended LE 7.5 mg non-responder group. Figure 2A illustrates the distribution of follicles across the three groups. Similarly, the mean endometrial thickness on the day of the trigger injection was 7.64±1.63 mm (range, 4 to 14) in the standard group, 7.69±1.56 mm (range, 5 to 12) in the extended LE 5 mg group, and 8.11±1.61 mm (range, 6 to 11) in the dose-escalated extended LE 7.5 mg group (Figure 2B). Ovulation was observed in 96.8% of patients in the standard group, 95% of patients in the extended LE 5 mg group, and 64.7% of patients in the LE 7.5 mg non-responder group (p=0.745) (Figure 2C).
Initial treatment outcomes in the standard, extended letrozole (LE) 5 mg, and extended LE 7.5 mg groups. (A) Distribution of the number of follicles (>14 to ≤18 mm). (B) Distribution of endometrial thickness (mm). (C) Ovulation rate.
5. Pregnancy outcomes
Patients with a positive pregnancy test (urine pregnancy test/β-hCG >5 mIU/mL) comprised 22.8% of the standard group, 15.6% of the extended LE 5 mg group, and 11.76% of the extended LE 7.5 mg non-responder group. Biochemical pregnancies (6%) and ectopic pregnancies (4%) were observed only in the standard group. Clinical pregnancy (i.e., a viable ongoing pregnancy at 10 weeks of gestation) was confirmed in 18.72% of patients in the standard group, 14.45% in the extended LE responder group, and 11.76% in the extended LE 7.5 mg non-responder group. The live birth rate was 16.89% in the standard group, 13.2% in the extended LE responder group, and 11.76% in the extended LE 7.5 mg non-responder group, with no statistically significant differences. Multiple pregnancies (10.8%) were observed only in the standard group (Table 3). Per-protocol analysis of clinical pregnancy rate at 10 weeks’ gestation between the standard group (41/219) and the LE extended group who received 5 mg for 10 days (12/83) revealed a p-values of 0.49 and a risk ratio (RR) of 1.295 (95% confidence interval [CI], 0.716 to 2.34). Comparison between the standard group and the extended LE 7.5 mg non-responder group (2/17) yielded a p-values of 0.744 and an RR of 0.159 (95% CI, 0.42 to 6.02), with no statistically significant difference.
Comparison of pregnancy outcomes between standard and extended letrozole protocols
6. Regression analysis
A binary logistic regression analysis was performed to investigate the possible confounding influences of age, BMI, AFC, and AMH on ovulation and clinical pregnancy at 10 weeks in women with PCOS receiving standard versus extended LE treatment regimens. In the standard LE group, BMI was identified as a significant confounder for ovulation (odds ratio [OR], 0.74; 95% CI, 0.60 to 0.91; p=0.004), indicating that higher BMI was negatively correlated with ovulation. However, this effect did not extend to clinical pregnancy at 10 weeks of gestation (p=0.075). Additional variables such as age, AFC, and AMH had no significant impact on outcomes. In the extended LE group, none of the variables exhibited a statistically significant impact on ovulation or clinical pregnancy. The analysis indicated a nearly significant relationship between age and ovulation (OR, 1.729; 95% CI, 0.962 to 3.107; p=0.067), but no consistent confounding factors were detected (Table 4).
Binary logistic regression with ovulation and clinical pregnancy at 10 weeks of gestation as dependent variables in the standard and extended letrozole groups
Discussion
This is the first retrospective study conducted to assess the efficacy of a novel stepwise intracycle extended and dose-escalated LE protocol combined with a low dose of gonadotropins in women with PCOS undergoing IUI. The investigation was performed in a tertiary care facility with a substantial population of patients with resistant PCOS. Consequently, the initial LE dosage was established at 5 mg daily for 5 days to increase the probability of OI (considered standard treatment). In patients who did not respond to this therapy, extended LE treatment was initiated at 5 mg/day for an additional 5 days (total of 10 days), and for those still not responding, the dosage was further escalated to 7.5 mg/day for a maximum of 5 days (range, 2 to 5). This protocol was based on the concept that increasing the LE dosage and extending the treatment duration within the same cycle could promote follicular development greater than 10 mm in highly resistant patients and delay or avoid an early switch to injectable gonadotropin therapies. In addition, the FSH window has been reported to play a crucial role in the selection of the dominant follicle [17]. Medication administered for a longer duration has been reported to extend the FSH window, inducing follicle growth in patients who initially did not respond to routine treatment [16].
In this study, extended LE treatment (5 mg for 7 to 10 days) resulted in follicular growth in 83 patients who did not respond to standard LE therapy, with ovulation in 95% and a clinical pregnancy rate of 14.4%. Subsequently increasing the LE dosage to 7.5 mg for the highly resistant non-responder group resulted in follicular growth with ovulation in 11 of the 17 patients, and two of these patients achieved a clinical pregnancy. The remaining six patients did not respond to the higher 7.5 mg dose. Similar findings have been reported with a two-step regimen using a duration-dependent protocol. In that study, 5 mg/day of LE was administered for 7 days, with extension to 10 days in subsequent cycles if ovulation was not achieved. The study reported an overall ovulation rate of 92.75% and a clinical pregnancy rate of 31.88%, with no significant differences between groups [16]. Furthermore, studies that utilized LE protocols with a short duration (LE, 5 mg/day for 5 days) and a long duration (LE, 2.5 mg/day for 10 days) have reported a greater pregnancy rate in the extended LE duration group [13,18]. Prior research has demonstrated the efficacy of extended LE therapy at a fixed dosage; the present study enhances this approach by integrating stepwise intracycle extended and dose-escalated LE, presenting a potential cost-effective alternative with favorable outcomes for highly resistant PCOS.
Studies have shown that patients with resistant PCOS usually have higher BMI, longer menstrual cycles, and higher AMH and AFC values, reflecting more severe hyperandrogenism and increasing resistance to ovarian stimulation [19,20]. This was reflected in the present study, in which the extended protocol group displayed statistically significant differences in these parameters.
Follicle development between the two regimens was also assessed. Similar monofollicular and multifollicular growth was observed in the standard and extended groups. Multiple pregnancies were observed only in the standard group, directly correlating with the higher percentage of cycles with at least four follicles (8.67%) developed. These results indicate that extended LE can promote follicular development without triggering an overstimulation response in women with highly resistant PCOS. A retrospective study on extended LE therapy reported similar findings, with a high percentage of monofollicular development and a low incidence of hyper-response [21].
In 2024, Dasgupta [15] conducted a sequential study on 50 LE-resistant patients using CC. For patients resistant to clomiphene, either gonadotropins or laparoscopic ovarian drilling was given. Patients who received gonadotropin therapy exhibited 100% ovulation; however, only 37.5% conceived. One patient developed ovarian hyperstimulation with 10 follicles, and another had triplets, leading to miscarriage [15]. In contrast, the present trial used a stepwise extended LE regimen within the same cycle for patients resistant to standard LE therapy (5 mg for 5 days), thereby reducing treatment delays and minimizing the risk of overstimulation.
A combination of LE and gonadotropins has been found to be superior in enhancing pregnancy and live birth rates in patients with infertility undergoing IUI compared to LE alone [22,23]. For OI, monofollicular growth is preferred, whereas in IUI settings, multifollicular growth (more than one mature follicle) is often targeted to increase treatment success [24]. Accordingly, in this study, once the dominant follicle reached >10 mm, a low dose of gonadotropin was administered. This approach ensured that the administration of gonadotropins did not influence the initial development of the dominant follicle, but rather supported its growth to raise the number of preovulatory follicles and potentially increase the pregnancy rate with IUI. A substantial body of evidence indicates that the combination of gonadotropins and oral ovulogens in IUI cycles can enhance multifollicular growth [25,26]. In a 2003 study by Healey et al. [27], a combination of LE and gonadotropins yielded an increased number of preovulatory follicles compared to gonadotropins alone. These results indicate that stepwise intracycle extended and dose-escalated LE offers significant advantages in patients with anovulatory PCOS.
This retrospective study was conducted at a tertiary-level fertility unit. The unit protocol is to discharge patients at 10 weeks of gestation for antenatal care elsewhere. Hence, detailed maternal and neonatal outcomes were not systematically recorded. A telephonic follow-up conducted post-delivery found that all infants born (n=50) to patients from this retrospective study cohort were reported to be healthy, with no congenital anomalies. However, future prospective studies are needed to evaluate the detailed safety outcomes of extended LE therapy with intracycle dose escalation.
In conclusion, the results of this study show that when first-line OI agents such as LE are ineffective in traditional protocol settings, newer optimized protocols can be clinically feasible in inducing follicular growth and reducing the need for gonadotropins as a second-line treatment. Further randomized trials comparing this stepwise protocol with traditional second-line gonadotropin regimens are needed to more fully evaluate its safety and efficacy.
Notes
Conflict of interest
No potential conflict of interest relevant to this article was reported.
Acknowledgments
The author thanks Dr. Sirisha Boddapati for manuscript support and the staff of the KIMS Fertility Center, KIMS Hospital, Secunderabad, for their cooperation and assistance.