The Impact of Late Luteinizing Hormone-Releasing Hormone Agonist Dosing on Testosterone Suppression in Patients with Prostate Cancer: An Analysis of United States Clinical Data
Abstract
Purpose:
We evaluated the timeliness of androgen deprivation therapy dosing, the impact of dosing nonadherence on testosterone, and the frequency of testosterone and prostate specific antigen testing in patients with prostate cancer.
Materials and Methods:
We retrospectively analyzed the records of 22,860 patients with prostate cancer treated with luteinizing hormone-releasing hormone agonists. Analyses were done using 2 definitions of month, including a 28-day month (late dosing after day 28, 84, 112 or 168) and an extended month (late after day 32, 97, 128 or 194) for 1, 3, 4 and 6-month formulations, respectively. The prevalence of late dosing, associated testosterone values, and the frequency of testosterone and prostate specific antigen testing were assessed. Statistical significance was assessed with the unpaired t-test.
Results:
Of the injections 84% and 27% were late for the 28-day and extended month analyses, respectively. For the 28-day month 60% and 29% of injections were late by more than 1 and more than 2 weeks, respectively. Of testosterone values 4% were greater than 50 ng/dl for early/on time injections using both definitions, and 15% and 27% were greater than 50 ng/dl when late, and for the 28-day month and the extended month, respectively. For early/on time vs late injections 22% vs 31% of testosterone values were greater than 20 ng/dl for the 28-day month and 21% vs 43% for the extended month. Mean testosterone was higher when late (49 ng/dl for 28-day month, 79 ng/dl for extended month) vs early/on time (both 21 ng/dl). Of the injections prostate specific antigen measurements were performed in 83% and testosterone assessment was done in only 13%.
Conclusions:
Luteinizing hormone-releasing hormone agonists were frequently (84%) administered later than the schedules used in pivotal trials. Nearly half of the late testosterone values for the extended month were greater than 20 ng/dl and mean testosterone was almost double the castration level. Elevated testosterone remained unidentified with infrequent testing.
| ADT |
androgen deprivation therapy |
| HPG |
hypothalamic-pituitary-gonadal |
| LHRH |
luteinizing hormone-releasing hormone |
| PCa |
prostate cancer |
| PSA |
prostate specific antigen |
| T |
testosterone |
Androgen deprivation therapy is the standard of care for advanced PCa, which is a hormone sensitive tumor, and LHRH agonists and antagonists are widely used to suppress serum T to castration levels.1 Therefore, T assessment is an appropriate marker of ADT effectiveness, as noted in drug package inserts and recommended by urology association guidelines. The current castration level is T less than 50 ng/dl but recent evidence suggests that less than 20 ng/dl is closer to what is seen following surgical castration.2 Maintaining T below the latter benchmark further delays disease progression and improves survival.3–7
In 2014 the EAU (European Association of Urology) guidelines updated target T during ADT to less than 20 ng/dl.8,9 The AUA (American Urological Association) has not yet adopted this target.10,11 However, recent draft guidance from the FDA (Food and Drug Administration) includes a recommendation to assess the achievement of T less than 20 ng/dl in registration trials for gonadotropin-releasing hormone analogues and include these data in product labels.12
ADT effectiveness in clinical practice relies on adherence to dosing schedules, correct mixing and administration, and periodic T monitoring. Adherence to cancer therapy schedules is critical as nonadherence is associated with treatment failure, decreased survival and the potential for greater health care utilization and costs.13–15 One source reported 56% adherence16 and a 10-database systematic review of factors influencing active cancer treatment revealed that adherence rates varied from 52% to 100% in patients 65 years old or older.13 To our knowledge no large-scale study has examined this for ADT and it appears that adherence rates of this frontline cancer therapy remain low.
Definitions of LHRH dosing schedules vary. Labels of FDA approved ADT drugs include dosing periods of calendar months, weeks or even days as derived from clinical studies (table 1). For reimbursement, payers may define a month as lasting 30 or 31 days. Adherence to schedules may be affected by health care delivery and patient influences. Delays in dosing may relate to rigid reimbursement policies denying early dosing, inconvenient or unavailable appointment scheduling and a shortage of clinic personnel compounded by increasing numbers of patients with PCa. Patient influences include scheduling challenges, difficulty with clinic proximity and transportation, and lack of understanding the importance of timely dosing. Additionally, health care providers and patients may be unaware of the potential consequences of even a short delay in dosing regardless of the ADT used.
| Drug (trade name) | Mo Definition (No. days) | Labeled Dosing Period | Prescribing Information Statement on T Monitoring |
| Leuprolide acetate: | |||
| Eligard® | 28 | 1, 3, 4 or 6 Mos | Response should be monitored by measuring serum T and PSA periodically |
| Lupron Depot® | 28 | 4, 12, 16 or 24 Wks | Periodic serum T and PSA monitoring is recommended, especially if anticipated clinical or biochemical response to treatment has not been achieved |
| Triptorelin (Trelstar®) | 28 | 4, 12 or 24 Wks | Response should be monitored by measuring serum T periodically or as indicated |
| Goserelin (Zoladex®) | 28 | 28 Days or 12 wks | No statement |
PSA levels in patients with PCa on and off ADT are monitored for a potential correlation with disease activity.17 However, because the primary goal of ADT is T reduction and low T correlates with improved disease specific survival,5 T is a more appropriate biomarker for evaluating ADT effectiveness. It is important to maintain low T to avoid escapes, during which LHRH agonist injection may cause a T surge (acute on chronic) as the HPG axis is no longer fully inhibited. Insufficient T suppression may have multiple causes, including late ADT dosing, a suboptimal mixing and/or administration technique, or ineffective treatment.
Despite the potential consequences of late dosing, to our knowledge no data are available on its impact on T suppression. This study investigated a large, single payer database, including access to electronic medical records and claims data, of patients with PCa who were receiving LHRH agonists. Study objectives included the timeliness of ADT dosing, the impact on T of nonadherence to dosing schedules and the frequency of T and PSA testing.
Materials and Methods
Study Design
We performed a retrospective analysis from January 2007 to June 2016 of 22,860 patients with PCa (ICD-9-185/10-C61) who received 2 or more injections of a LHRH agonist, ie goserelin, leuprolide acetate (intramuscular or subcutaneous) or triptorelin, and who had 1 or more measurement of PSA or T. Clinical data were collected from an integrated source including claims from hospitals, multispecialty practices, small group practices and physician offices. Patients were under the care of urologists, oncologists or physicians of other specialties.
All LHRH agonist products were grouped together for analysis. Formulation refers to 1, 3, 4 and 6-month doses regardless of the molecule and the delivery system. Data on patients who received intermittent ADT were included as the impact of delayed dosing would not have depended on whether the regimen was intermittent or continuous.
Month Definitions
Two definitions of month allowed detailed investigation into the frequency of delayed dosing and impact on T levels (fig. 1). Definition 1 was a 28-day month, a strict definition derived from pivotal trials in which 28 days defined a month. Late was defined as dosing after day 28, 84, 112 and 168 for the 1, 3, 4 and 6-month formulations, respectively (table 2). Early/on time injections were defined as dosing on or before day 28, 84, 112 or 168. Definition 2 was an extended month. In clinical practice ADT dose scheduling is frequently based on calendar months or longer periods under the assumption that efficacy extends beyond the labeled dosing periods. This second definition added approximately 4 days per month to the 28-day month interval for each formulation based on independent clinical advice. Late referred to dosing after day 32, 97, 128 or 194 for each of the 4 formulations, respectively. We performed analyzed early/on time and late dosing for each definition and the impact on T.
Figure 1. Extent of late dosing used in 1-month formulation analyses
| Mo | 28-Day Mo | Extended Mo | ||
| On Time Days | Late Days | On Time Days | Late Days | |
| 1 | 28 | 29 or More | 32 | 33 or More |
| 3 | 84 | 85 or More | 97 | 98 or More |
| 4 | 112 | 113 or More | 128 | 129 or More |
| 6 | 168 | 169 or More | 194 | 195 or More |
Analysis Methodology
All analyses were done using descriptive statistics. The unpaired 2-sample t-test was applied to investigate between group differences with p <0.05 considered significant. Four formulations were analyzed individually and pooled.
Late Dosing Prevalence
The prevalence of late dosing was the total number of late injections divided by the total number of injections. First injections in each patient were excluded.
Late Dosing Effect on T Suppression
We evaluated T values for all T tests that reported a date of draw and used the unit ng/dl. T tests prior to a subsequent injection and after the previous injection served as the individual data points to analyze the impact of delayed dosing. Due to the surge in T after the first dose of LHRH agonists, the analyses of the 1-month formulation included and excluded these levels. The percent of tests that exceeded 50 and 20 ng/dl as well as mean T values were evaluated to determine the magnitude of effect. Values less than 0.01 ng/dl (below the limit of detection and undetectable) were entered as 0.009 ng/dl since excluding them could have introduced substantial bias.
Testosterone and Prostate Specific Antigen Testing Frequency
Testing frequency was determined as the number of injections with an associated T or PSA test divided by the total number of injections. PSA tests required a blood draw date and a value in ng/ml.
Results
Patient Demographics, and Testosterone and Prostate Specific Antigen Test Frequency
The 22,860 patients received a total of 85,030 LHRH agonist injections from January 2007 to June 2016, including 10,669 injections for the 1-month, 30,311 for the 3-month, 28,777 for the 4-month and 15,273 for the 6-month formulations. Demographics were well balanced across the 4 formulations. Mean age was 73 to 75 years, and 73% to 78% of participants were Caucasian, 10% to 15% were African American, 1% to 3% were Asian and 9% to 13% were another race or ethnicity (table 3). Although patients on intermittent ADT could not be identified, less than 4% of injections were more than 6 months late.
| Formulation | ||||
| 1-Mo | 3-Mo | 4-Mo | 6-Mo | |
| No. pts | 3,175 | 7,065 | 7,359 | 5,246 |
| Mean age (range) | 73.3 (43–89) | 73.0 (26–89) | 74.5 (43–89) | 74.6 (20–88) |
| % Race (No.): | ||||
| Caucasian | 76.8 (2,439) | 73.6 (5,200) | 73.2 (5,390) | 77.8 (4,080) |
| African American | 11.2 (356) | 10.5 (741) | 14.6 (1,075) | 11.8 (617) |
| Asian | 1.4 (44) | 3.3 (231) | 1.1 (84) | 1.1 (57) |
| Other | 10.6 (336) | 12.6 (893) | 11.0 (810) | 9.4 (492) |
Of ADT injections 83% were associated with a PSA measurement and only 13% (11,284) were associated with a T assessment. Frequency was similar across the formulations, including 60% for PSA, while for T the frequency was 12% for the 1-month formulation, and 86% and 15% for the 3-month, 84% and 11% for the 4-month, and 89% and 14% for the 6-month formulations, respectively.
Late Dosing
Prevalence
For the 28-day month definition 84% of subsequent ADT injections were late, including 60% more than 1 week and 29% more than 2 weeks late. For the extended month definition 27% of injections were late, including 13% more than 1 week late and 9% more than 2 weeks late. Late injections were common across all formulations (fig. 2).
Figure 2. Proportion of LHRH agonist injections administered late, pooled and by formulation. Yellow bars indicate 28-day month. Blue bars indicate extended month.
Impact on Testosterone Suppression
Only 4% of early/on time injections were associated with T more than 50 ng/dl using both definitions, while 15% of 28-day month and 27% of extended month injections were associated with T more than 50 ng/dl when late (fig. 3). The impact of late dosing on T greater than 20 ng/dl was greater than the impact on T greater than 50 ng/dl. For early/on time vs late, 22% vs 31% of T were greater than 20 ng/dl using 28-day month definition and 21% vs 43% using extended month definition (fig. 4). Furthermore, when dosing was late by more than 2 weeks, the negative impact on T suppression was larger with a 26% incidence of T greater than 50 ng/dl and a 41% incidence greater than 20 ng/dl for the 28-day month, and 35% and 50%, respectively, for the extended month definitions.
Figure 3. Proportion of LHRH agonist injections with T greater than 50 ng/dl. A, 28-day month. B, extended month.
Figure 4. Proportion of LHRH agonist injections with T greater than 20 ng/dl. A, 28-day month. B, extended month.
Each definition demonstrated higher mean T levels for late vs early/on time injections. For the 28-day month early/on time injections resulted in a mean ± SD T of 21 ± 56 ng/dl. Late injections resulted in a mean T of 49 ± 107 ng/dl (table 4). For injections more than 2 weeks late mean T was 74 ± 131 ng/dl. For extended month injections mean T increased from 21 ± 59 ng/dl when early/on time to 79 ± 135 ng/dl when late. Mean T increased to 98 ± 147 ng/dl for injections more than 2 weeks late.
| Mo Definition | Early or On Time | Late |
| 28-Day: | ||
| Mean ± SD ng/dl | 21 ± 56 | 49 ± 107 |
| Median ng/dl (IQR) | 11 (7–20) | 13 (7–26) |
| Extended: | ||
| Mean ± SD ng/dl | 21 ± 59 | 79 ± 135 |
| Median ng/dl (IQR) | 11 (7–20) | 19 (9–71) |
For each definition the proportion of T greater than 50 ng/dl and greater than 20 ng/dl as well as mean T significantly differed between early/on time and late injections (each p <0.01). Trends were consistent across all 4 formulations. For T evaluation within the first month of treatment with the 1-month formulation, results were consistent with the analyses excluding these values, suggesting that the T surge had ended.
Discussion
ADT aims to achieve and maintain T below the castration level (ideally 20 ng/dl or less) to delay disease progression and prolong survival. High T negatively impacts survival,3,6–8,18 highlighting the significance of failure to adequately suppress T. Historically surgical castration was performed to achieve ADT but chemical castration has emerged as the treatment of choice due to reversibility and fewer associated risks.19 Rising PSA may reflect tumor progression to castration resistance11 but late ADT dosing or issues with the mixing and/or administration technique may be implicated. With rising PSA it is critical to ensure that T is below castration levels to prevent an erroneous diagnosis of castration resistance.
The main objective of this study was to evaluate dosing adherence and its impact on T suppression. Following a review of the literature, to our knowledge this is the first study to analyze a substantial amount of quantitative data on patients with PCa. Findings refer primarily to metastatic PCa for which ADT is accepted as first line treatment.20 However, findings may also be relevant in patients at high risk with nonmetastatic disease.
ADT injections were frequently (84%) administered beyond the strict 28-day month definition used in pivotal trials. While for the extended month definition (about 4 extra days per month) that may be more representative of clinical practice, delays were still frequent with more than 25% of doses administered late. Delays substantially impacted T suppression with almost half of tests showing a value greater than 20 ng/dl compared to a fifth of tests for dosing on time. The impact was greatest when injections were more than 2 weeks late with mean T more than fourfold higher (98 vs 21 ng/dl) than for early/on time injections. Mean T levels when ADT was dosed early/on time were 21 ng/dl for each month definition, confirming that LHRH agonists suppress the gonadal release of T in a way similar to that of surgical castration.
The low adherence to ADT dosing and the impact on efficacy would not be acceptable for many routinely used drugs. Late dosing of anticoagulants, antihypertensives and statins could increase blood clotting, blood pressure or cholesterol with serious clinical consequences. Ineffective dosing of chemotherapy drugs in oncology settings could negatively impact patient outcomes, including survival. Nonadherence to ADT dosing risks the restoration of HPG axis function, increased T and the potential for acute-on-chronic T surges with subsequent doses.
This study guides physicians to interpret T tests based on injection timeliness. If dosing is early/on time and T is more than 50 or 20 ng/dl, treatment may be considered ineffective. However, if dosing is late and T is more than 50 or 20 ng/dl, treatment may not be effective beyond the dosing period. Physicians should consider aligning practice processes and educating staff and patients to improve dosing adherence.
Only 13% of ADT doses in this study had an associated T measurement, implying that inadequate T suppression is often unidentified. Current practice relies on PSA as a surrogate of ADT efficacy. However, since the primary goal of ADT is T reduction to castration levels, efficacy should be validated by T measurement.21 This is increasingly important with the advent of new therapies of castration resistant disease which require a correct diagnosis before initiation. LHRH agonist package inserts recommend periodic T assessment to monitor suppression (table 1). Operationally appropriate T testing would be simple and convenient if the assay was ordered concurrently with PSA via simultaneous blood draws. T should be measured using validated laboratory instruments and sensitive assay technologies to accurately measure very low T levels.22,23 Modern technologies with a lower limit of detection have revealed much lower T than earlier methods.24
The study has several limitations. The retrospective design is more appropriate for hypothesis generation than for providing confirmatory evidence. In the absence of other relevant publications, analyses of alternative databases would have helped to corroborate the findings. However, the large size and diversity of the data set suggest that our results likely represent current clinical experience.
Although the study was not randomized or controlled, it would be impractical and unethical to purposely delay ADT to confirm inferior clinical outcomes. Because analyses were done across multiple institutions, standardizing instrumentation and T assay methodology was not feasible, but the large size of the data set should have reduced any impact on conclusions.
Finally, as T was monitored infrequently, it is possible that patients who were not responding well from a clinical or PSA biomarker standpoint were more likely to undergo T measurement, potentially resulting in bias. However, as T suppression differed for early/on time vs late dosing, the same trend would likely have been observed with consistent T monitoring.
Time for T to rise following ADT cessation varies between individuals and depends on patient age, body mass index, ADT duration and baseline T.25–27 Therefore, rapid increases in T above the castration level following delays in dosing may have implications for cancer progression.6,7,28,29 More research into the effects of delayed or missed doses on disease progression and outcomes would be valuable. Other recommendations are to investigate patient outcomes, including the impact on mortality and time to onset of castration resistance, and differences between LHRH agonist therapies in the maintenance of T suppression. Long-term followup would also be insightful since prolonged ADT may lead to chronic T suppression due to the burnout of Leydig cells and the HPG axis.25
Considering the established clinical benefits of effective T suppression, clinicians should ensure that patients undergo optimal ADT dosing schedules. Monitoring T and PSA would allow efficacy and changes in PCa management to be reviewed, which should improve clinical outcomes. It is particularly important to educate urology care providers on the importance of adherence to ADT dosing schedules since practices are relying more on advanced practice providers to monitor and care for patients who receive chronic ADT.30
Conclusions
This study of ADT dosing in clinical practice demonstrates that 84% of LHRH agonists were administered later than scheduled (28 days or multiples thereof) in pivotal trials with consequent T elevations. Nearly half of the late T tests (extended month definition) showed more than 20 ng/dl T with mean T levels almost twice the castration level. Many patients with elevated T remain unidentified due to infrequent testing. It is highly likely that increasing adherence to ADT dosing schedules would improve T suppression in patients with PCa and potentially improve clinical outcomes, including delaying disease progression. Clinicians should consider modifying practice procedures to maximize the benefits of ADT in patients with PCa before determining that progression to the next stage has occurred.
Acknowledgment
Dr. Ingrid Koo, Xelay Acumen Group, Inc., provided editorial support.
REFERENCES
- 1. National Comprehensive Cancer Network®: NCCN Guidelines Prostate Cancer Guidelines. Available at https://jnccn.org/view/journals/jnccn/17/5/article-p479.xml. Accessed September 18, 2019. Google Scholar
- 2. : Reassessment of the definition of castrate levels of testosterone: implications for clinical decision making. Urology 2000; 56: 1021. Crossref, Medline, Google Scholar
- 3. : Testosterone in prostate cancer: the Bethesda consensus. BJU Int 2012; 110: 344. Crossref, Google Scholar
- 4. : Incomplete testosterone suppression with luteinizing hormone-releasing hormone agonists: does it happen and does it matter?BJU Int 2012; 110: E500. Crossref, Google Scholar
- 5. : Nadir testosterone within first year of androgen-deprivation therapy (ADT) predicts for time to castration-resistant progression: a secondary analysis of the PR-7 trial of intermittent versus continuous ADT. J Clin Oncol 2015; 33: 1151. Google Scholar
- 6. : Testosterone levels in patients with metastatic prostate cancer treated with luteinizing hormone-releasing hormone therapy: prognostic significance?BJU Int 2010; 105: 648. Google Scholar
- 7. : Redefining clinically significant castration levels in patients with prostate cancer receiving continuous androgen deprivation therapy. J Urol 2007; 178: 1290. Link, Google Scholar
- 8. : Androgen-targeted therapy in men with prostate cancer: evolving practice and future considerations. Prostate Cancer Prostatic Dis 2019; 22: 24. Google Scholar
- 9. : Guidelines on Prostate Cancer. Prostate Cancer (text update April 2014). Eur Urol 2014; 65: 467. Google Scholar
- 10. : Castration-resistant prostate cancer: AUA guideline amendment 2018. J Urol 2018; 200: 1264. Link, Google Scholar
- 11. : Design and end points of clinical trials for patients with progressive prostate cancer and castrate levels of testosterone: recommendations of the Prostate Cancer Clinical Trials Working Group. J Clin Oncol 2008; 26: 1148. Crossref, Medline, Google Scholar
- 12. Food and Drug Administration: Advanced Prostate Cancer: Developing Gonadotropin-Releasing Hormone Analogues. Guidance for Industry, 2019. Available at https://www.fda.gov/media/129027/download. Accessed September 18, 2019. Google Scholar
- 13. : Factors influencing adherence to cancer treatment in older adults with cancer: a systematic review. Ann Oncol 2014; 25: 564. Google Scholar
- 14. : Nonadherence to imatinib adversely affects event free survival in chronic phase chronic myeloid leukemia. Am J Hematol 2011; 86: 471. Google Scholar
- 15. : Completion of adjuvant radiation therapy among women with breast cancer. Cancer 2008; 113: 22. Google Scholar
- 16. : Patient compliance in treatment of prostate cancer with luteinizing hormone-releasing hormone (LHRH) agonist. Urology 1993; 42: 533. Crossref, Medline, Google Scholar
- 17. : PSA and beyond: alternative prostate cancer biomarkers. Cell Oncol (Dordr) 2016; 39: 97. Google Scholar
- 18. : Testosterone suppression in the treatment of recurrent or metastatic prostate cancer—a Canadian consensus statement. Can Urol Assoc J 2018; 12: 30. Google Scholar
- 19. : Metabolic complications of androgen deprivation therapy for prostate cancer. J Urol 2009; 181: 1998. Link, Google Scholar
- 20. : New considerations for ADT in advanced prostate cancer and the emerging role of GnRH antagonists. Prostate Cancer Prostatic Dis 2013; 16: 7. Crossref, Google Scholar
- 21. : Serum testosterone level is a useful biomarker for determining the optimal treatment for castration-resistant prostate cancer. Urol Oncol 2019; 37: 485. Google Scholar
- 22. : Accuracy of serum luteinizing hormone and serum testosterone measurements to assess the efficacy of medical castration in prostate cancer patients. J Biomed Sci 2017; 24: 81. Google Scholar
- 23. : Discordance between testosterone measurement methods in castrated prostate cancer patients. Endocr Connect 2019; doi: 10.1530/EC-18-047. Crossref, Google Scholar
- 24. : Serum testosterone assays—accuracy matters. J Clin Endocrinol Metab 2004; 89: 520. Google Scholar
- 25. : Factors associated with testosterone recovery after androgen deprivation therapy in patients with prostate cancer. Investig Clin Urol 2018; 59: 18. Google Scholar
- 26. : Testosterone recovery profiles after cessation of androgen deprivation therapy for prostate cancer. J Sex Med 2019; 16: 872. Google Scholar
- 27. : Failure to achieve castrate levels of testosterone during luteinizing hormone releasing hormone agonist therapy: the case for monitoring serum testosterone and a treatment decision algorithm. J Urol 2000; 164: 726. Link, Google Scholar
- 28. : Long-term suppression of luteinizing hormone, follicle-stimulating hormone and testosterone by daily administration of leuprolide. J Urol 1987; 137: 57. Link, Google Scholar
- 29. : Clinical effects of gonadotropin-releasing hormone analogue in metastatic carcinoma of prostate. Urology 1985; 25: 106. Crossref, Google Scholar
- 30. and Spitz A: Use of advanced practice providers as part of the urologic healthcare team. Curr Urol Rep 2015; 16: 62. Crossref, Medline, Google Scholar
The corresponding author certifies that, when applicable, a statement(s) has been included in the manuscript documenting institutional review board, ethics committee or ethical review board study approval; principles of Helsinki Declaration were followed in lieu of formal ethics committee approval; institutional animal care and use committee approval; all human subjects provided written informed consent with guarantees of confidentiality; IRB approved protocol number; animal approved project number.
Supported by Tolmar Pharmaceuticals, Inc.
No direct or indirect commercial, personal, academic, political, religious or ethical incentive is associated with publishing this article.
