Healthcare resource use in advanced prostate cancer patients treated with docetaxel

Abstract

Objective:

Although the treatment of metastatic castrate-resistant prostate cancer (mCRPC) has improved with newer therapies, there is little understanding how these therapies have impacted resource use and associated expenditures; available estimates are dated. The current study examined contemporary healthcare utilization and associated costs for mCRPC patients and how these measures changed over time.

Methods:

This retrospective cohort analysis used medical and pharmaceutical insurance claims data from a large non-payer-owned integrated claims database of US commercial insurers. Amongst all patients with a prostate cancer diagnosis (n = 256,464), those with ≥ 1 docetaxel claim (docetaxel cohort, n = 3642) were identified as mCRPC patients. Within the docetaxel cohort, an additional 6-months follow-up cohort (n = 2862) was identified, i.e., patients with at least 6 months of follow-up after the first docetaxel claim. Resource utilization and costs were identified for all-cause hospitalizations, emergency room (ER) visits, physician visits and ambulatory visits, and prostate cancer-related prescription treatments.

Results:

Significant increases in the mean per-patient-per-month (PPPM) count for the docetaxel cohort were observed for all medical resources measured (hospitalizations and ER, physician, and ambulatory visits) in the post-docetaxel period compared with the pre-docetaxel period (p < 0.0001); similar significant increases were observed for the 6-months follow-up cohort in the last 6 months (prior to lost to follow-up date) compared with the period preceding the last 6 months (p < 0.0408 ambulatory visits, p < 0.0001 all other resources). Total docetaxel cohort costs (mean [standard deviation]) rose from an average PPPM cost of US$2593 (3208) in the pre-docetaxel period to US$5847 (6990) in the post-docetaxel period (p < 0.0001); each of the individual resources measured (hospitalization, all healthcare visits, and prescription costs) demonstrated significant increases (p < 0.0001).

Limitations:

Retrospective study design.

Conclusions:

This large database analysis showed a significant increase in use of healthcare resources and associated costs among mCRPC patients following first-line docetaxel treatment.

Keywords::

• Metastatic castrate-resistant prostate cancer
• Chemotherapy
• Health outcomes
• Health economics
• Cost
• Retrospective cohort study

Introduction

Prostate cancer is the most commonly diagnosed solid tumor amongst males in the US and Europe1,2. Although early-stage prostate cancer has an excellent prognosis (100% 5-year survival rate), the outlook for patients with advanced disease is discouraging, with <32% expected to live 5 years or more3. Advanced prostate cancer (APC) is characterized by either rising prostate-specific antigen with nodal/visceral lesions or by growth of measurable disease4 and is also referred to as metastatic castrate-resistant prostate cancer (mCRPC) because it progresses despite castrate levels of testosterone.

In 2004, docetaxel in combination with prednisone was approved for patients with mCRPC5, as it demonstrated a significant, albeit modest, survival advantage over mitoxantrone plus prednisone, the previous standard regimen6. This approval signaled a paradigm shift in treatment algorithms and consequently healthcare resource utilization for mCRPC. While several new agents have been recently approved for mCRPC, current clinical guidelines7 recommend docetaxel plus prednisone as the first-line standard of care (SOC) for patients with symptomatic mCRPC; the new alternatives are recommended either for asymptomatic mCRPC (sipuleucel-T) or as second-line treatment (abiraterone acetate, cabazitaxel).

Prostate cancer is associated with substantial treatment costs and overall healthcare resource use8–11. However, data quantifying these costs in a large patient population with advanced disease are sparse. A small study published in 2002 demonstrated the economic burden of mCRPC, but this analysis was conducted prior to the approval of docetaxel therapy and hence did not capture the costs of docetaxel treatments11. A larger study published in 2010 reported per-patient-per-month (PPPM) costs of nearly US$1800 for patients with CRPC12. However, this study was conducted from 2001–2007, meaning that docetaxel was not available for much of this time frame; not surprisingly, <25% of that patient sample received docetaxel, presumably as first-line SOC for advanced disease. Therefore, despite docetaxel currently being the current first-line SOC for symptomatic mCRPC, information on the economic impact of this treatment option remains very limited.

We used data from a non-payor owned, comprehensive, integrated claims database to indirectly identify a large cohort of patients with mCRPC, based on standard diagnosis codes for prostate cancer and with more than one claim for docetaxel therapy. In order to better understand the health-economic impact of mCRPC, we analyzed patterns of healthcare resource utilization and costs among this large group of commercially-insured patients receiving docetaxel for APC.

Materials and methods

Data source and sample selection

This was a retrospective cohort analysis using medical and pharmaceutical insurance claims data obtained from the PharMetrics® IMS LifeLink^TM Health Plan claims database (IMSIMS Health, Danbury, CT). The PharMetrics database is the largest non-payer-owned integrated claims database of commercial insurers in the US, and is used for many retrospective, real-world pharmacoeconomic studies13–17. At the time of the analysis, this de-identified database included medical and pharmacy claims for more than 55 million unique members from more than 90 health plans countrywide. The database is representative of the national commercially insured population and collects a variety of demographic measures such as age, gender, and plan type. Available data also include patient diagnoses (International Statistical Classification of Diseases, Ninth Revision [ICD-9] codes), commercial medical claims from all healthcare sites (i.e., hospital, emergency room [ER], outpatient setting, etc.), and retail and mail-order pharmacy claims.

In this study, the following groups of patients were identified from the PharMetrics database within the 7-year time period of January 1, 2003 to December 31, 2009:

1. Total prostate cancer population: patients with a prostate cancer diagnosis (ICD-9 code 185.X) who were enrolled in the database for ≥12 months.

2. Docetaxel cohort: patients within the total prostate cancer population who had ≥1 claim for docetaxel (HCPCS code J9170).

3. 6-months follow-up cohort: patients within the docetaxel cohort who had a follow-up period of ≥6 months from first docetaxel claim to either loss to follow-up or October 31, 2009 (whichever came first).

Resource use evaluation

Resource use measures

The following resource use measures were assessed: hospitalizations, ER visits, physician visits, and ambulatory visits (all-cause for each); and prostate cancer-related prescription treatments.

The docetaxel cohort was analyzed for resource use and costs incurred during two sequential time periods (see Figure 1): (1) the period prior to the first docetaxel claim (pre-docetaxel period); and (2) the follow-up period from the first docetaxel claim until loss to follow-up or censor date (post-docetaxel period). Additionally, during the post-docetaxel period, resource and costs incurred were analyzed for (1) the 6 months preceding loss to follow-up or censor date (defined as the ‘last 6 months period’) and (2) the period starting from the first docetaxel claim until the beginning of the ‘last 6 months period’ (defined as the ‘prior to last 6 months period’). Lastly, the short-term follow-up cohort—defined as those with less than 6 months follow-up after the first docetaxel claim—was analyzed for resource use and costs incurred from the first docetaxel claim until loss to follow-up or censor date.

Figure 1.  Study time periods.

Number of events

Incidence of all-cause hospitalizations, ER visits, physician visits, and ambulatory visits was derived from the database using American Medical Association place-of-service codes. All-cause hospitalization was captured on a monthly basis, so that multiple hospitalizations for the same event and the same patient within a 1-month time frame were counted only once. Rate per person-months for hospitalization, ER visits, physician visits, and ambulatory visits were calculated as: rate/person-month = total events/cumulative follow-up (in months). In addition, PPPM counts for hospitalization, ER visits, physician visits, and ambulatory visits were calculated as: PPPM count = total events/total follow-up duration (in months) for each patient. PPPM counts were presented as mean (standard deviation [SD]) for each type of visit.

Exposure to the various prostate cancer-related prescription treatments was calculated based on the presence of J and/or NDC codes for docetaxel, other chemotherapies (i.e., estramustine, mitoxantrone, and vinorelbine), androgen deprivation therapy (ADT), ketoconazole, prednisone, bisphosphonates, and (non-opioid and opioid) analgesics. Because patients generally received multiple prostate cancer treatments, we specified the following definitions to allow for a more meaningful hierarchical analysis:

1. Docetaxel use: docetaxel claim with or without claims for any other prostate cancer therapies.

2. Other chemotherapy use: claim for any single-agent or combination chemotherapy regimen (excluding patients with docetaxel claims).

3. Androgen deprivation therapy use: claim for any ADT with or without claim(s) for ketoconazole and/or prednisone (excluding patients with chemotherapy claims).

4. Ketoconazole use: claim for ketoconazole with or without claim for prednisone (excluding patients with claims for chemotherapy or ADT).

5. Prednisone use: claim for prednisone only.

Cost analysis

PPPM costs of all-cause hospitalizations, ER visits, physician visits, and ambulatory visits were calculated as PPPM cost = total cost/PPPM count. PPPM costs account for the varying durations of follow-up captured in the database after the first docetaxel claim. Costs retrieved were so-called allowed costs, that is, the amount allowed by the plan for a particular service, which is typically the paid amount + any member liability (i.e., copay, deductible, or coinsurance). PPPM costs were presented as mean (SD) for each type of visit.

PPPM cost was also calculated for total prescription treatments (i.e., docetaxel + other chemotherapies + ADT + ketoconazole + prednisone + analgesics + bisphosphonates). Subsequently, total prostate cancer-related treatment PPPM costs were calculated as: Total PPPM cost = PPPM costs of hospitalizations, ER visits, physician visits, ambulatory visits, and total prescription treatments. Total costs for each treatment were based on all claims for the treatment by each patient. Treatment switches and costs associated with the new treatment were also tracked for each patient.

Statistical analysis

The analyses were largely descriptive. Numbers and proportions were calculated for treatment claims and healthcare utilization measures. Mean and SD were calculated for continuous demographic variables (e.g., age), healthcare resource utilization, and costs. Chi-square and t-tests were used to compare proportions/rates and means between the pre- and post-docetaxel periods, respectively. Chi-square and t-tests were also used to compare proportions/rates and means between the ‘prior to the last 6 months’ and the ‘last 6 months periods’. Statistical significance was defined at the level of 5% (α = 0.05). All statistical analysis was conducted using SAS® software, version 9.1 (SAS Institute, Inc., Cary, NC).

Results

Patient disposition

Using the criteria detailed above, 256,464 patients from the PharMetrics database were identified for the total prostate cancer population, of whom 3642 patients (1.4%) met criteria for inclusion into the docetaxel cohort (Figure 2). In the docetaxel cohort, the mean duration from time of enrollment to first docetaxel claim was 10.29 (4.04) months, with a median of 12 months; the mean duration of follow-up after first docetaxel claim was 15.82 (13.79) months, with a median of 12 months. The docetaxel cohort was further sub-divided into the 6-months follow-up cohort (n = 2862) and the short-term follow-up cohort (n = 780). The 6-months cohort had a mean duration of follow-up from first docetaxel claim until the start of the last 6 months of 13.25 (13.66) months, with a median of 9 months.

Figure 2.  Patient disposition.

Patient characteristics

Table 1 shows that baseline characteristics were similar in the total prostate cancer population and the docetaxel cohort; mean patient age was 69.20 (11.53) years and 69.62 (10.20) years, respectively. Approximately 70% of both patient populations had commercial insurance, ∼20% were enrolled in Medicare-Risk/Medicare Advantage or Medicaid (with self/commercial insurance), and ∼10% were self-insured.

Table 1.  Baseline characteristics of the total prostate cancer population and the docetaxel cohort.

Characteristic, n (%)	Total prostate cancer population (n = 256,464)	Docetaxel cohort (n = 3642)
Age, mean (SD)	69.20 (11.53)	69.62 (10.20)
<40 years, n (%)	1329 (0.5)	4 (0.1)
40–54 years, n (%)	20,055 (7.8)	229 (6.3)
55–64 years, n (%)	71,443 (27.9)	946 (26.0)
65–74 years, n (%)	81,557 (31.8)	1261 (34.6)
≥75 years, n (%)	82,080 (32.0)	1202 (33.0)
Insurance type
Commercial, n (%)	179,724 (70.1)	2542 (69.8)
Medicare-Risk/Advantage,  Medicaid, n (%)	48,796 (19.0)	754 (20.7)
Self-insured, n (%)	26,848 (10.5)	334 (9.2)
Unknown, n (%)	1096 (0.4)	12 (0.3)

SD, standard deviation.

Resource use

Docetaxel cohort

In the docetaxel cohort, utilization of all medical resources measured (all-cause hospitalization, ER visits, physician visits, and ambulatory visits) was increased in the post-docetaxel period compared with the pre-docetaxel period, as measured by rate per person-month (Figure 3). When mean PPPM count was analyzed for these same medical resources, significant increases were observed in the post-docetaxel period across each measure (p < 0.0001; Table 2). For instance, the PPPM average for ambulatory visits, which exhibited the highest frequency among all resource categories in the pre-docetaxel period, increased by 56% (2.38 [1.58] vs 3.72 [2.14], p < 0.0001) and was the most frequently used resource in the post-docetaxel period.

Figure 3.  Resource utilization (rate per person-month) in the docetaxel cohort (n = 3642) across pre- and post-docetaxel periods.

Table 2.  Resource utilization in the docetaxel cohort (n = 3642) during the pre- and post-docetaxel periods. Per-patient-per-month (PPPM) counts of all hospitalizations and healthcare visits.

Resource utilization	PPPM count
	Pre-docetaxel period	Post-docetaxel period	p
All-cause hospitalization, mean (SD)	0.05 (0.10)	0.12 (0.19)	<0.0001
Emergency room visits, mean (SD)	0.07 (0.14)	0.15 (0.23)	<0.0001
Physician visits, mean (SD)	2.11 (1.47)	3.35 (2.17)	<0.0001
Ambulatory visits, mean (SD)	2.38 (1.58)	3.72 (2.14)	<0.0001

SD, standard deviation.

Among patients in the docetaxel cohort, the mean and median numbers of docetaxel prescriptions were 10.05 (11.68) and 6, respectively. The average duration of docetaxel treatment was 5.63 (5.24) months, with a median of 4 months; this mean duration is in agreement with the recommended docetaxel SOC regimen of six cycles (3 weeks on and 1 week off). Docetaxel re-treatment, defined as patients receiving ≥ 4 months of docetaxel with a break of ≥3 months prior to re-initiation of docetaxel, was observed in 264 (7.2%) patients.

Prescription prostate cancer-related therapies used in the docetaxel cohort are presented in Table 3. Nearly two-thirds of patients received ADT prior to docetaxel, but ≤7% had other chemotherapy claims (primarily mitoxantrone) during this same time period. With the exception of ketoconazole and ADT, the use of prescription prostate cancer-related treatments increased from the pre-docetaxel period to the post-docetaxel period; specifically, usage of prednisone, analgesics, and bisphosphonates rose by 76.5%, 18.2%, and 70.4%, respectively. The use of other chemotherapy was 3.6-times greater (24.7% vs 6.8%) in the post-docetaxel compared with the pre-docetaxel period. Approximately 42% of patients in the docetaxel cohort had exposure to analgesics at the same time or within a month before or after their first docetaxel claim.

Table 3.  Resource utilization in the docetaxel cohort (n = 3642) during the pre- and post-docetaxel periods. Treatment exposure to prostate cancer-related prescriptions.

Prescription treatment	Pre-docetaxel period n (%)	Post-docetaxel period n (%)
Docetaxel	0 (0)	3642 (100)
Other chemotherapy	249 (6.8)	898 (24.7)
Androgen deprivation therapy	2381 (65.4)	2113 (58.0)
Ketoconazole	364 (10.0)	218 (6.0)
Prednisone	759 (20.8)	1340 (36.8)
Analgesics	1832 (50.3)	2166 (59.5)
Bisphosphonates	1212 (33.3)	2065 (56.7)

6-months follow-up cohort

In the 6-months follow-up cohort, utilization of all medical resources, as measured by PPPM count, increased significantly in the last 6 months compared with the period preceding the last 6 months: hospitalizations more than tripled (p < 0.0001), ER visits more than doubled (p < 0.0001), physician visits increased by 20.1% (p < 0.0001), and ambulatory visits increased by 2.9% (p = 0.0408; Table 4). The proportions of patients within the 6-months cohort utilizing these medical resources also increased in the last 6 months (Figure 4): hospitalizations more than doubled, as did ER visits, and nearly all of the patients in this cohort (over 93%) had physician and/or ambulatory visits during the last 6 months of follow-up.

Figure 4.  Resource utilization (percentage of patients) in the 6-months follow-up cohort (n = 2862) during the period prior to the last 6 months and during the last 6 months period.

Table 4.  Resource utilization in the 6-months follow-up cohort (n = 2862) during the period prior to the last 6 months and the last 6 months period. Per-patient-per-month (PPPM) counts of all hospitalizations and healthcare visits.

Resource utilization	PPPM count
	Prior to last 6 months period, Mean (SD)	Last 6 months period, Mean (SD)	p
All-cause hospitalization	0.04 (0.11)	0.14 (0.21)	<0.0001
Emergency room visits	0.06 (0.16)	0.16 (0.24)	<0.0001
Physician visits	2.69 (2.08)	3.23 (2.52)	<0.0001
Ambulatory visits	3.43 (2.46)	3.53 (2.77)	0.0408

SD, standard deviation.

Cost analysis

Docetaxel cohort

The distribution of total prostate cancer-related costs among the docetaxel cohort was US$100.5 million in the pre-docetaxel period; this included all hospitalization charges, all healthcare visits, and all prostate cancer-related prescription costs (Figure 5). This total cost increased by 163% to US$264.8 million in the post-docetaxel period. Ambulatory costs were the greatest expenditure in the pre-docetaxel period (43.5% of the total); however, prostate cancer-related prescription costs increased from 16.2% of the total during the pre-docetaxel period to 34.2% in the post-docetaxel period, thus capturing the largest single proportion of the total cost.

Figure 5.  Distribution of total prostate cancer-related costs in the docetaxel cohort across pre- and post-docetaxel periods (n = 3642).

PPPM analysis of costs incurred by the docetaxel cohort also demonstrated an increase in expenditures in the post-docetaxel period. Table 5 shows that total costs rose from an average PPPM allowed cost of US$2593 (3208) in the pre-docetaxel period to US$5847 (6990) in the post-docetaxel period (p < 0.0001). Total prescription costs accounted for 15.8% of overall PPPM costs in the pre-docetaxel phase, a proportion that rose to 33.2% in the post-docetaxel phase. This higher cost during the latter time period was primarily due to docetaxel, which had a mean PPPM allowed cost of US$1411 (1458), equivalent to 72.7% of the total prescription cost.

Table 5.  Per-patient-per-month (PPPM) allowed costs in the docetaxel cohort (n = 3642) during the pre- and post-docetaxel periods.

PPPM allowed costs	Pre-docetaxel period, Mean (SD), US$	Post-docetaxel period, Mean (SD), US$	p
Total costs	2593 (3208)	5847 (6990)	<0.0001
All-cause hospitalization	686 (2040)	1800 (5447)	<0.0001
Emergency room visits	29 (91)	60 (162)	<0.0001
Physician visits	344 (699)	600 (1095)	<0.0001
Ambulatory visits	1125 (1551)	1444 (2147)	<0.0001
Total prescription treatment costs	409 (523)	1941 (1808)	<0.0001
Prostate cancer treatment	254 (341)	1623 (1581)	<0.0001
Analgesics	20 (91)	57 (226)	<0.0001
Bisphosphonates	135 (290)	261 (386)	<0.0001

SD, standard deviation.

Total prescription treatments = prostate cancer treatments + analgesics + bisphosphonates; prostate cancer treatments = docetaxel, other chemotherapies, androgen deprivation therapy, ketoconazole, and prednisone.

6-months follow-up cohort

The average total PPPM cost of the last 6 months among the 6-months follow-up cohort was US$5293 (7128), which was significantly higher than the average PPPM cost of US$3834 (4781) in the period prior to the last 6 months (p < 0.0001), as shown in Table 6. Aside from prescription treatments, the costs for each medical resource increased significantly in the last 6 months (p < 0.0001). Whereas prescription treatment and ambulatory visits were responsible for the majority of costs in the former period (US$1305 and US$1278, respectively), hospitalizations were the greatest single expense in the last 6 months of follow-up, for a PPPM cost of US$1957 (37.0% of the total PPPM cost). In the short-term cohort, the average total PPPM cost was US$7909 (11,119). Resource utilization across all categories was higher compared with those in the 6-months cohort (data not shown).

Table 6.  Per-patient-per-month (PPPM) allowed costs in the 6-months follow-up cohort (n = 2682) during the period prior to the last 6 months and the last 6 months period.

Resource utilization	PPPM allowed costs
	Prior to last 6 months period, Mean (SD), US$	Last 6 months period, Mean (SD), US$	p
Total costs	3834 (4781)	5293 (7128)	<0.0001
All-cause hospitalization	749 (2541)	1957 (5216)	<0.0001
Emergency room visits	38 (253)	68 (182)	<0.0001
Physician visits	464 (1125)	611 (1295)	<0.0001
Ambulatory visits	1278 (2342)	1534 (2623)	<0.0001
Total prescription treatment costs	1305 (1976)	1122 (1615)	<0.0001

SD, standard deviation.

Total prescription treatments = prostate cancer treatments + analgesics + bisphosphonates; prostate cancer treatments = docetaxel, other chemotherapies, androgen deprivation therapy, ketoconazole, and prednisone.

Discussion

Although mCRPC is a significant problem in the developed world and its treatment is undeniably costly8–11,18, previous studies quantifying healthcare resource use and cost were conducted before the 2004 introduction of docetaxel as the SOC for mCRPC11,12. The current analysis is the first to examine contemporary healthcare utilization and associated costs for mCRPC patients treated with current first-line SOC using a large real-world database, with particular attention paid to how these measures change over time. Our results were consistent with previously published observations suggesting that resource use and costs associated with the management of mCRPC are substantial8–11,18,19, and clearly indicated that disease progression (defined here as exposure to docetaxel therapy along with prostate cancer diagnosis) of mCRPC following first-line docetaxel treatment leads to a significant increase in the use of healthcare resources and associated costs. Although increases were observed in all categories of healthcare resources, the greatest additional expenditure after docetaxel exposure was noted in hospitalizations and prescription costs. Rising prescription costs were driven primarily by docetaxel expenses; however, the use of most other prostate cancer-related prescription therapies also increased in patients following docetaxel exposure. Of note, based on these data, the rate of re-treatment with docetaxel appears to be very low.

Results from the 6-months follow-up cohort, in which healthcare resource use from the post-docetaxel period was divided into claims recorded before and after the last 6 months of follow-up, were similar to those from the docetaxel cohort: the latter time period saw a significantly greater use of resources and significantly higher associated costs. However, whereas prescription costs increased significantly in the docetaxel cohort overall during the post-docetaxel period, the 6-months cohort actually exhibited a decrease in prescription expenditures during the last 6 months.

Pain is a significant issue among advanced prostate cancer patients. In the current study, ∼42% of prostate cancer patients in the docetaxel cohort used analgesics within 1 month before or after their first docetaxel claim. This number is similar to the proportion of patients from the landmark docetaxel study who had pain at baseline (i.e., 45%)6; however, those patients had to have a specific pain level (≥2 on a scale of 0–5) or a specific level of daily analgesic use (≥10, on a scale in which 1 represented one non-narcotic dose and 4 represented one narcotic dose). Since the current study measured any amount of baseline analgesic use, this patient cohort may have used fewer analgesics overall than the patients from the docetaxel study.

While this is the first large-scale medical resource use analysis in this setting to use real-world data, the study has certain limitations due to its retrospective design and the type of data available from the database. For instance, patients reliant solely on Medicare (i.e., without supplemental insurance) are not captured in the PharMetrics database, so the results of the current study may not be generalizable to Medicare-only prostate cancer patients.

In addition, this analysis was hampered by the lack of a specific ICD-9 code for mCRPC—as a matter of fact, most published studies characterize this disease state in variable terms20. We used an a priori definition framed within the context of the methodology used and the available data to identify patients with mCRPC, i.e., the occurrence of at least one docetaxel claim among those with a prostate cancer diagnosis. While this was an indirect method to categorize the population of interest, our approach is supported by the fact that systemic chemotherapy is recommended strictly for patients with mCRPC and that docetaxel is the current first-line SOC7. Fewer than 7% of patients in the docetaxel cohort were exposed to other chemotherapies prior to docetaxel, while nearly one-quarter of patients received other chemotherapies in the post-docetaxel period, likely as salvage therapy. Taken together, it is thus reasonable that in the vast majority of patients in our sample, docetaxel was used as systemic chemotherapy for treatment of mCRPC. Patients post-docetaxel therapy also had an increased use of prostate cancer-related prescription therapies overall, which indirectly further confirms the accuracy of our patient selection criteria.

Another limitation of the study was that, given the nature of the database, we were unable to identify the precise reasons for the increased medical resource use observed in the post-docetaxel period. This could be the result of actual disease progression, an increase in treatment-related toxicities, or a number of other factors. For instance, a recent US study showed that CRPC patients treated by oncologists, especially those using chemotherapy, had higher total and prostate cancer-related healthcare cost than CRPC patients treated by urologists21. When comparing CRPC patients not using chemotherapy, the treatment costs were similar, suggesting that severity of illness can be a key factor in driving cost of care21 and, hence, would be an important variable to consider in future healthcare cost analyses.

Finally, this study could not account for the use of two new second-line options for the treatment of mCRPC that have recently been approved by the US Food and Drug Administration—cabazitaxel and abiraterone acetate—because these were not available during the study analysis period. Beetsch et al.22 recently demonstrated that, with the introduction of new therapies, the treatment algorithms available for patients and physicians, as well as payment strategies, become increasingly complex. Newer treatments offer significant benefits in terms of survival and quality-of-life, but they are also more costly compared to older hormonal therapies22. Future studies should examine the impact of these new therapies on healthcare-related resource use and costs in patients with mCRPC and also explore ways of quantifying other benefits and incorporating them into the cost structure.

Conclusions

Using a large claims database, we found a significant increase in healthcare resource use and associated cost among APC patients after docetaxel treatment, especially during the last 6 months of their follow-up. New and improved therapeutic options are needed to improve survival and decrease resource use. Further analyses are warranted to determine the specific reasons for the disproportionate resource use and costs of mCRPC patients treated with docetaxel during later time periods.

Transparency

Declaration of funding

This study was funded by Janssen Global Services.

Declaration of financial/other relationships

Maneesha Mehra and Ravinder Dhawan are full-time employees of Janssen Global Services. Ying Wu was a full-time Janseen Global Services employee at the time the analysis was conducted and the manuscript was written.

Acknowledgments

This study was funded by Janssen Global Services. Writing assistance was provided by D. Wolf of PAREXEL, and was funded by Janssen Global Services.