Abstract
Background:
Bone modifying agents (BMAs) prevent skeletal related events among patients with metastatic, castration-resistant prostate cancer (mCRPC) involving bone and prevent osteoporotic fractures among patients at high risk. BMA utilization for patients with mCRPC has not been well quantified.
Methods:
We used linked SEER registry and Medicare claims data. We included men diagnosed with stage IV prostate adenocarcinoma during 2007–2015, aged >= 66 at diagnosis, with sufficient continuous enrollment in Medicare Parts A, B, and D, who received androgen deprivation therapy. We limited to those who subsequently received a CRPC-defining treatment (CDT). We identified patients with evidence of bone metastasis using claims. Our primary outcome was receipt of a BMA (zoledronic acid or denosumab) within 180 days of initiating CDT.
Results:
Among 1292 included patients, 1034 (80%) had bone metastasis. BMA use within 180 days of initiating CDT was higher among patients with bone metastases than those without (705/1034 [68%] vs 56/258 [22%]). Among patients without bone metastasis, those with high osteoporotic fracture risk were more likely than those without to receive a BMA (OR=2.48, 95%CI 1.17, 5.29); however, only 26% of patients with high fracture risk received a BMA. Among patients who received BMAs, most (62%) first initiated them >90 days before initiating CDT.
Conclusions:
Two-thirds of patients with mCRPC and bone metastases received BMAs within 180 days after initiating CDT. A greater proportion of patients without bone metastasis may warrant BMA therapy for osteoporotic fracture prevention. Some patients with bone metastasis may be able to delay BMA initiation until CRPC.
Keywords: bone modifying agents, guideline-based care, high-value care, practice patterns, prostate cancer, supportive care
Introduction
Prostate cancer causes the death of over 34,000 people annually, making it one of the most deadly cancers among US men. Prostate cancer commonly metastasizes to bone, causing skeletal-related events (SREs): pathologic fracture, spinal cord compression, hypercalcemia, and bone pain requiring intervention. SREs affect 30–50% of patients with bone metastases,1,2 causing substantial loss of function and quality of life.3
The bone-modifying agents (BMAs) zoledronic acid and denosumab prevent SREs and are recommended by NCCN guidelines for patients with metastatic, castration-resistant prostate cancer (mCRPC) with bone involvement.4 Additionally, BMAs are recommended to prevent osteoporotic fractures among patients with clinical osteoporosis,4,5 as defined by dual-energy x-ray absorptiometry (DEXA) measurement or by a personal history of prior osteoporotic fracture.6 Osteoporotic fracture prevention is particularly relevant to prostate cancer, because its older patient population has a higher prevalence of osteoporosis and osteopenia7 and because androgen deprivation therapy (ADT) further accelerates bone mineral density loss.8 BMAs cause toxicities including renal failure, electrolyte abnormalities such as hypocalcemia, and osteonecrosis of the jaw.
Despite the demonstrated benefit from guideline-concordant BMA therapy, little is known about real-world treatment patterns. Studies of other cancers suggest BMAs may be underused,9–11 but it remains unclear whether this is also the case for prostate cancer. Several studies have measured BMA use in prostate cancer, with estimates ranging from 20% to 61% of patients receiving BMAs.12–14 These studies were limited by the inclusion of patients with metastatic, castration sensitive prostate cancer (mCSPC) – for whom BMAs are not recommended unless high osteoporotic fracture risk is also present – potentially biasing the measurement of utilization among mCRPC patients.
An additional important question regarding BMA therapy is the relative utilization of zoledronic acid and denosumab. Denosumab is marginally more effective than zoledronic acid in SRE prevention,15–17 but has greater incidence of some toxicities (severe electrolyte abnormalities)18 and greater financial cost ($53 per dose of zoledronic acid vs. $2,321 for denosumab),19 causing it to be non-cost-effective at conventional thresholds.20 Denosumab is less nephrotoxic, however, so it may be preferred for patients with underlying renal disease.21 Whether real-world utilization reflects these financial and patient-level considerations is unknown.
We aimed to characterize BMA utilization among patients with mCRPC, with attention to patient characteristics (such as bone metastasis) and provider factors. We hypothesized greater utilization among patients with bone metastases or high fracture risk, in line with guideline recommendations. We also hypothesized greater relative use of denosumab for patients with renal disease and for those treated in the physician office setting, as previously observed among patients with mCSPC.22
Methods
Patient population
We used linked SEER-Medicare data from 2007–2015. We included fee-for-service Medicare beneficiaries diagnosed with stage IV prostate adenocarcinoma whose first cancer diagnosis was prostate cancer (or second cancer diagnosis if the first occurred >=3 years prior), aged ≥66 at diagnosis, not diagnosed at death, and had continuous enrollment in Medicare Parts A and B, and not Medicare Advantage, for >=240 days prior to month of diagnosis and through the 180 day outcome period. We required continuous enrollment in Medicare Part D from the month of diagnosis through the outcome period. We included patients with a first claim for hormonal therapy (ADT or antiandrogen) or orchiectomy in proximity to prostate cancer diagnosis.
The analytic cohort was the subset of these patients who subsequently received a CRPC-defining treatment (CDT) (Supplementary Figure S1). Briefly, these included either an agent which, during the study period, was indicated only for mCRPC (abiraterone acetate, enzalutamide, mitoxantrone, docetaxel, cabazitaxel, other cytotoxic agents, radium-223, sipuleucel-T), or a switch in therapy if the initial therapy may have been indicated for mCSPC.
We used a claims-based algorithm to identify presence of bone metastases. Presence of bone metastasis was defined by 1 or more inpatient claims with a primary or secondary diagnosis of bone metastases OR 2 or more outpatient claims for treatments, diagnostic procedures, or evaluation & management with diagnosis codes for bone metastases, occurring between 8 months prior to diagnosis and initiation of CDT. In a sensitivity analysis, we compared restrictive (requiring 2 or more outpatient claims >30 days apart, and excluding diagnostic procedures) and permissive (requiring only 1 or more outpatient claims) versions of the algorithm.
Outcome definition
The primary outcome was any claim for a BMA (zoledronic acid or denosumab) within 180 days after initiation of CDT. Although oral bisphosphonates may be appropriate for prevention of osteoporotic fractures, we did not include these drugs because their use was very rare in our sample (<11 individuals). Because patient factors may cause either more rapid or delayed BMA initiation to be appropriate for some patients, we included pre-specified secondary endpoints of BMA therapy within 90 and 365 days. Importantly, patients were assessed for this outcome regardless of whether they received BMAs prior to CDT.
Among patients who met the primary outcome, we further characterized the timing of BMA initiation. Patients were grouped with respect to whether the first BMA claim occurred: 1) from diagnosis to 90 days prior to CDT (during which time patients were likely still to have CSPC), 2) during the 90 days prior to CDT, or 3) after CDT.
To infer the intended indication of BMA therapy (SRE prevention vs. osteoporotic fracture prevention), we measured dose strength and timing. Receipt of a denosumab dose of 120mg or greater, or receipt of >1 zoledronic acid doses within a 5 month period, were defined as SRE prevention; all lower or less-frequent doses were defined as osteoporotic fracture prevention.
Patient characteristics and potential confounders
We identified patient characteristics which may be associated with receipt of BMA therapy. These included age, calendar year of diagnosis, race/ethnicity, kidney disease, modified Charlson index,23 ZIP code-level median income (American Community Survey, 2008–2012), and time duration from diagnosis to CDT. Provider characteristics of interest included treatment billing location (hospital outpatient, HO vs. physician office, PO), defined hierarchically by 1) location of BMA administration among those that received BMA, 2) location of CDT administration among those whose CDT was physician-administered, 3) location of majority of Evaluation & Management claims.24–26 We identified involvement of endocrinology by E&M claims with a provider of that specialty.
We defined a set of patient conditions which may warrant BMA therapy for osteoporotic fracture prevention. These included osteoporosis, osteopenia, prior bone fractures, and other SREs (cord compression, bone radiation, bone surgery); hypercalcemia was also included as it is an appropriate BMA indication even though it does not indicate higher fracture risk.
Statistical analysis
To assess associates of BMA utilization, we performed multivariable logistic regression using the full cohort and the subsets with and without bone metastasis. To assess associates of denosumab utilization, we performed multivariable logistic regression on the subset that received any BMA therapy (either drug) and was diagnosed after denosumab market entry (2010 onwards).
Two sided p-values of <0.05 were considered significant. All analyses were conducted using SAS version 9.4. The Memorial Sloan Kettering IRB found this research exempt under 45 CFR 46.104(d)(4).
Code availability
Code available in supplemental materials.
Results
We identified 1,292 patients diagnosed with stage IV prostate cancer from 2007–2015 who subsequently received CDT (Supplementary Figure S1). Of these, 1,034 (80.0%) had evidence of bone metastases before initiation of CDT (Table 1). 733 (56.7%) were treated in the PO setting, and 559 (43.3%) in HO. Median age at diagnosis was 75, and median time to CDT was 15 months (IQR 10, 26). As a proportion of CDT, docetaxel decreased during the study period while abiraterone acetate and enzalutamide increased (Supplementary Figure S2).
Table 1: Patient characteristicsa.
Overall primary cohort, and subdivided by bone metastasis status (by primary method; see Table S1).
| Characteristic | Primary cohort | Bone metastasis | No bone metastasis |
|---|---|---|---|
| N | N | N | |
| 1292 | 1034 | 258 | |
| Age at Diagnosis | |||
| Median (IQR) | 75 (71, 81) | 75 (71, 81) | 76 (71, 82) |
| Mean (SD) | 76 (7) | 76 (7) | 76 (7) |
| Race, N (%) | |||
| White | 1049 (81%) | 847 (82%) | 202 (78%) |
| Black | 109 (8%) | 81 (8%) | 28 (11%) |
| Other | 134 (10%) | 106 (10%) | 28 (11%) |
| Year of Diagnosis, N (%) | |||
| Before 2010 | 337 (26%) | 260 (25%) | 77 (30%) |
| 2010 or 2011 | 272 (21%) | 215 (21%) | 57 (22%) |
| 2012 or later | 683 (53%) | 559 (54%) | 124 (48%) |
| Treatment Location, N (%) | |||
| Physician Office | 733 (57%) | 564 (55%) | 169 (66%) |
| Hospital Outpatient | 559 (43%) | 470 (45%) | 89 (34%) |
| Time to CDT, months | |||
| Median (IQR) | 15 (10, 26) | 14 (10, 24) | 20 (13, 32) |
| Mean (SD) | 20 (15) | 19 (14) | 25 (17) |
| CDT, N (%) | |||
| Docetaxel | 376 (29%) | 329 (32%) | 47 (18%) |
| Cabazitaxel | 0 (0%) | 0 (0%) | 0 (0%) |
| Sipuleucel-T | 108 (8%) | 94 (9%) | 14 (5%) |
| Abiraterone acetate | 399 (31%) | 351 (34%) | 48 (19%) |
| Enzalutamide | 191 (15%) | 157 (15%) | 34 (13%) |
| Radium-223 | 13 (1%) | *c | *c |
| First-generation antiandrogen | 197 (15%) | 89 (9%) | 108 (42%) |
| Other chemotherapy | 29 (2%) | *c | *c |
| Endocrinology visit, N (%) | |||
| After prostate cancer diagnosis | 175 (14%) | 140 (14%) | 35 (14%) |
| After initiation of CDT | 47 (4%) | *c | *c |
| Comorbid Conditions, N (%) | |||
| 0 | 859 (66%) | 683 (66%) | 176 (68%) |
| 1 | 214 (17%) | 165 (16%) | 49 (19%) |
| 2+ | 219 (17%) | 186 (18%) | 33 (13%) |
| Predicted probability of dependence in ADLsb | |||
| Median (IQR) | 7% (5%, 11%) | 7% (5%, 11%) | 7% (4%, 12%) |
| Mean (SD) | 11% (12%) | 11% (12%) | 11% (12%) |
| Survival time (months) | |||
| Median (IQR) | 28 (18, 42) | 26 (18, 40) | 35 (22, 51) |
| Mean (SD) | 33 (20) | 31 (19) | 40 (22) |
| History of renal disease, N (%) | |||
| No | 1080 (84%) | 859 (83%) | 221 (86%) |
| Yes | 212 (16%) | 175 (17%) | 37 (14%) |
| ZIP code median income, N (%) | |||
| Less than $50k | 458 (35%) | 370 (36%) | 88 (34%) |
| $50k to 54999 | 108 (8%) | 84 (8%) | 24 (9%) |
| $55k to 59999 | 95 (7%) | *c | *c |
| $60k to 69999 | 189 (15%) | 149 (14%) | 40 (16%) |
| $70k or more | 413 (32%) | 333 (32%) | 80 (31%) |
| Unknown | 29 (2%) | *c | *c |
ADLs; activities of daily living; IQR, inter-quartile range; SD, standard deviation; CDT, CRPC-defining treatment.
Faurot algorithm
suppressed to avoid re-identification as some cell sizes <11.
Overall, 761 (58.9%) patients received a BMA within 180 days after CDT (Table 2). 703 (54.4%) received BMA therapy within 90 days after CDT, and 810 (62.7%) within 365 days. Grouping by evidence of bone metastases, 705 (67.5%) patients with bone metastases received BMA within 180 days, as did 56 (21.6%) without bone metastases. Using the restrictive algorithm to identify bone metastases, 912 (70.6%) had bone metastases, of which 634 (69.5%) received BMAs (Supplementary Table S1). Using the permissive algorithm, 1,094 (84.7%) had bone metastases, of which 730 (66.7%) received BMAs.
Table 2: Primary and secondary outcomesa.
Receipt of bone modifying agents within 180, 90, and 365 days of diagnosis after initiation of CRPC-defining treatment for all patients and stratified by presence of bone metastases.
| Characteristic | Overall (%) | Bone metastasis (%) | No bone metastasis (%) | |
|---|---|---|---|---|
| Total | 1292 | 1034 | 258 | |
| 180 days | Any BMA | 761 (59%) | 705 (68%) | 56 (22%) |
| Denosumab | 430 (33%) | 395 (38%) | 35 (14%) | |
| Zoledronic acid | 331 (26%) | 310 (30%) | 21 (8%) | |
| Secondary outcomes | ||||
| 90 days | Any BMA | 703 (54%) | 662 (64%) | 41 (16%) |
| Denosumab | 393 (30%) | 369 (36%) | 24 (9%) | |
| Zoledronic acid | 310 (24%) | 293 (28%) | 17 (7%) | |
| 365 days | Any BMA | 810 (63%) | 742 (72%) | 68 (26%) |
| Denosumab | 469 (36%) | 425 (41%) | 44 (17%) | |
| Zoledronic acid | 341 (26%) | 317 (31%) | 24 (9%) | |
BMA, bone modifying agent.
Overall BMA utilization was stable across the study period. Among patients with bone metastases who began in CDT in 2008 or later, the proportion of patients who met the primary outcome ranged between 61–79% across the study period (Figure 1). The relative utilization of zoledronic acid decreased (79% in 2010 vs. 18% in 2016) while denosumab increased (0% in 2010 vs. 48% in 2016).
Figure 1: Proportion of patients with bone metastasis who received a bone modifying agent within 180 days after initiating CRPC-defining treatment, over time.
Calendar year indicates initiation of CRPC-defining treatment, rather than year of prostate cancer diagnosis.
Osteoporosis/osteopenia was the most common condition which may warrant BMA therapy for osteoporotic fracture prevention, present in 584 (56.5%) of patients with bone metastases and 106 (41.1%) of patients without bone metastases (Supplementary Table S2). Among patients with bone metastases, 784 (75.8%) had at least one such condition, of whom 535 (68.2%) received BMA therapy. Among patients without bone metastases, 160 (62.0%) had at least one condition and 41 (25.6%) received BMA therapy.
Among the 761 patients who received BMA therapy (for any indication) within 180 days after CDT, most (473, 62.2%) initiated BMA more than 90 days prior to CDT; only 175 (23.0%) first initiated BMA subsequent to CDT, and 113 (14.8%) did so during the 90 days prior to CDT (Table 3). Among the 761 patients, 485 (63.7%) had one or more conditions that may warrant BMA therapy; this was similar across patients who initiated BMA therapy at different times. Assessing the subset of patients who initiated BMA more than 90 days prior to CDT, the large majority of BMA therapy was dosed for SRE prevention rather than osteoporotic fracture prevention regardless of the presence of bone metastases or indicators of high fracture risk (Supplementary Table S3).
Table 3: Timing of BMA initiation among patients who received BMA within 180 days after CRPC-defining treatmenta.
Any bone fracture includes pathologic fracture, spine or hip fracture, and non-pathologic non-spine/hip fracture. Osteoporosis, osteopenia, and spine or hip fracture were measured over a patient’s entire history prior to BMA initiation. Pathologic fracture, non-pathologic non-spine/hip fracture, other skeletal-related events (cord compression, bone surgery, bone radiation), and hypercalcemia were measured from the time of prostate cancer diagnosis to BMA initiation.
| Interval during which BMA was initiated | # | % | Osteoporosis or osteopenia (%) | Any Bone fracture (%) | Other SRE (%) | Hypercalcemia (%) | Any condition (%) |
|---|---|---|---|---|---|---|---|
| Overall | 761 | 366 (48%) | 119 (16%) | 198 (26%) | 13 (2%) | 485 (64%) | |
| After prostate cancer diagnosis and before 90 days prior to CRPC date | 473 | 62% | 233 (49%) | 71 (15%) | 105 (22%) | <11 (<2%) | 296 (63%) |
| Within 90 days prior to CRPC date | 113 | 15% | 53 (47%) | 17 (15%) | 36 (32%) | <11 (<10%) | 77 (68%) |
| After CRPC date | 175 | 23% | 80 (46%) | 31 (18%) | 57 (33%) | <11 (<6%) | 112 (64%) |
| WITH bone metastasis | 705 | 335 (48%) | 113 (16%) | 182 (26%) | 12 (2%) | 446 (63%) | |
| After prostate cancer diagnosis and before 90 days prior to CRPC date | 461 | 65% | 225 (49%) | 71 (15%) | 102 (22%) | <11 (<2%) | 287 (62%) |
| Within 90 days prior to CRPC date | 107 | 15% | 48 (45%) | 17 (16%) | 35 (33%) | <11 (<10%) | 72 (67%) |
| After CRPC date | 137 | 19% | 62 (45%) | 25 (18%) | 45 (33%) | <11 (<8%) | 87 (64%) |
| WITHOUT bone metastasis | 56 | 31 (55%) | <11 (<20%) | 16 (29%) | <11 (<20%) | 39 (70%) | |
| After prostate cancer diagnosis and before 90 days prior to CRPC date | 18b | 32%b | <11 | <11 | <11 | <11 | <11 |
| Within 90 days prior to CRPC date | <11 | <11 | <11 | <11 | <11 | ||
| After CRPC date | 38 | 68% | 18 (47%) | <11 (<31%) | 12 (32%) | <11 (<31%) | 25 (66%) |
BMA, bone modifying agent; SRE, skeletal related event.
Combined total for both categories is reported to prevent re-identification due to cell sizes <11
Among patients with bone metastases, a longer time interval between diagnosis and CDT was associated with lower odds of BMA therapy (OR 0.61, 95%CI 0.46, 0.82) (Table 4). Among patients without bone metastases, presence of a condition that may warrant BMA therapy was associated with increased BMA therapy (OR 2.48, 95%CI 1.17, 5.29). No other factor was associated with BMA therapy. The only factors associated with denosumab use (vs. zoledronic acid) were later calendar year and age >=80 (Supplementary Table S4). HO vs. PO location was not associated with BMA therapy or denosumab use.
Table 4: Factors associated with receipt of any bone modifying agent within 180 days of CRPCa.
“Any BMA condition” indicates the presence of one or more of: osteoporosis, osteopenia, or spine or hip fracture (occurring any time prior to CRPC-defining treatment); pathologic fracture, non-pathologic non-spine/hip fracture, other skeletal-related events (cord compression, bone surgery, bone radiation), and hypercalcemia (occurring after prostate cancer diagnosis and prior to CRPC-defining treatment).
| Effect | Overall | Bone metastasis | No bone metastasis |
|---|---|---|---|
| OR (95% CI) | OR (95% CI) | OR (95% CI) | |
| HO location | Reference | Reference | Reference |
| PO location | 0.89 (0.7, 1.12) | 1.02 (0.78, 1.34) | 0.73 (0.37, 1.41) |
| Year of diagnosis | |||
| Before 2010 | Reference | Reference | Reference |
| 2010–2011 | 1.14 (0.81, 1.59) | 1.15 (0.77, 1.71) | 1.05 (0.39, 2.82) |
| 2012 or later | 1.16 (0.87, 1.54) | 1.02 (0.73, 1.43) | 2.05 (0.92, 4.58) |
| Time to CDT greater than median | 0.54 (0.42, 0.69) | 0.61 (0.46, 0.82) | 0.79 (0.39, 1.58) |
| Age <80 | Reference | Reference | Reference |
| Age >=80 | 0.8 (0.62, 1.03) | 0.8 (0.59, 1.09) | 1.02 (0.49, 2.1) |
| Race | |||
| White | Reference | Reference | Reference |
| Black | 0.72 (0.46, 1.11) | 0.65 (0.39, 1.09) | 1.48 (0.51, 4.25) |
| Other | 1.09 (0.75, 1.59) | 1.21 (0.76, 1.91) | 0.82 (0.28, 2.44) |
| Comorbid conditions | |||
| None | Reference | Reference | Reference |
| 1 | 0.97 (0.7, 1.33) | 1.14 (0.77, 1.69) | 0.63 (0.26, 1.48) |
| 2+ | 0.85 (0.61, 1.2) | 0.79 (0.54, 1.16) | 0.73 (0.25, 2.18) |
| History of kidney disease | 1.01 (0.72, 1.42) | 1.06 (0.71, 1.56) | 0.56 (0.18, 1.78) |
| Any BMA condition | 1.55 (1.19, 2.01) | 1.1 (0.8, 1.52) | 2.48 (1.17, 5.29) |
| ZIP code median income | |||
| <$50,000 | Reference | Reference | Reference |
| $50,000 to 54,999 | 0.76 (0.49, 1.17) | 0.7 (0.42, 1.17) | 1.45 (0.52, 4.04) |
| $55,000 to 59,999 | 0.77 (0.48, 1.22) | 0.89 (0.51, 1.56) | 0.39 (0.10, 1.54) |
| $60,000 to 69,999 | 0.92 (0.64, 1.32) | 1.12 (0.72, 1.75) | 0.45 (0.15, 1.39) |
| >=$70,000 | 0.74 (0.56, 0.99) | 0.71 (0.51, 0.99) | 0.70 (0.32, 1.53) |
OR, odds ratio; CI, confidence interval; HO, hospital outpatient; PO, physician office; CDT, CRPC-defining treatment.
Discussion
For CRPC, NCCN Guidelines recommend BMA therapy for SRE prevention among patients with bone metastasis and osteoporotic fracture prevention among high-risk patients.4 We found that over two-thirds of patients with CRPC and bone metastasis received BMA therapy, suggesting that patients with mCRPC generally receive guideline-concordant BMA therapy.
However, approximately one-third (32%) of patients with bone metastases did not receive a BMA. It is unclear whether – or to what degree - this represents underuse. Some patients may be ineligible based on comorbidities, or decline BMA therapy when it is recommended. Other patients (eg., those with solitary bone metastasis) may have low SRE risk, and physicians may decide against BMA therapy in such cases.27
Besides SRE prevention, BMAs are also recommended (at less frequent and/or lower dosing) for the prevention of osteoporotic fractures among those at high risk. This is most relevant to mCRPC patients without bone involvement, who do not already warrant BMA therapy for SRE prevention. Among patients without bone metastases, we found that while those with conditions that may warrant BMA therapy were more likely to receive a BMA (Table 4), a substantial portion of these patients did not (Supplementary Table S2). Furthermore, among patients with bone involvement and one or more conditions that may warrant BMA therapy, the large majority of BMA therapy administered prior to initiation of CDT is dosed for SRE prevention rather than osteoporotic fracture prevention (Supplementary Table S3), even though the higher SRE-prevention dosing would not be recommended until castration resistance. Taken together, these results suggest that BMA utilization for osteoporotic fracture prevention may warrant greater attention by providers.
Different factors were associated with BMA therapy among patients with vs. without bone metastases (Table 4). Among those with bone metastases, a shorter duration of time from diagnosis to CDT was associated with increased BMA use. This may be because these patients have more aggressive disease or a greater metastatic burden, and are appropriately prioritized for BMA therapy based on higher SREs risk. Among patients without bone metastases, the presence of conditions which may warrant BMA therapy for osteoporotic fracture prevention was associated with BMA therapy, also likely reflecting prioritization of high risk patients. Notably, treatment location was not associated with BMA therapy among CRPC patients, in contrast to prior findings of greater use in the PO location among CSPC patients, for whom BMA therapy may constitute overuse.22 Further research is needed to understand why treatment location may be associated with BMA overuse but not with appropriate use in the mCRPC setting.
The proportion of patients with mCRPC and bone metastasis who received BMA therapy was relatively unchanged across the study period (Figure 1), even as initial therapy for mCRPC shifted from docetaxel to abiraterone acetate and enzalutamide (Supplementary Figure S2) and BMA therapy shifted from zoledronic acid to denosumab, potentially due to clinical trial results suggesting superiority of denosumab or perceived ease of administration.
In most cases, BMA therapy was initiated for SRE prevention before it was likely to have been clinically indicated. Even allowing for the possibility that BMA therapy may be initiated prior to CDT in response to clinical CRPC (our rationale for including the time period of 90 days prior to CDT as a separate window for BMA initiation), the majority (65%) of patients with bone metastases who received BMA therapy had already initiated it while their disease was castrate sensitive (Table 3). Though most of these patients also had conditions which may warrant BMA therapy for osteoporotic fracture prevention, the BMA dosing they received suggest that the intent of therapy was SRE prevention rather than osteoporotic fracture prevention (Supplementary Table S3). This is in line with prior findings of substantial BMA use for CSPC in both US and European settings,22,28 and suggests the potential to reduce treatment costs and toxicity by delaying initiation of BMAs until castration resistance, or by applying osteoporotic fracture prevention dosing where appropriate.
Given the small incremental benefit of denosumab compared to zoledronic acid, the relative cost-effectiveness of these two agents appears to be sensitive to their pricing. A cost-effectiveness analysis of denosumab found that it was not cost-effective compared to zoledronic acid at US prices, even before availability of generic zoledronic acid.20 Another US-based analysis found that denosumab was cost-effective,29 but has been criticized for methodologic assumptions which resulted in an inflated estimate of denosumab’s clinical benefit.20,30 In contrast, a Netherlands-based study, where denosumab and zoledronic acid were similarly priced (€486 and €451, respectively), found denosumab to be cost-effective.31 We observed that denosumab quickly overtook zoledronic acid as a proportion of BMA therapy following its 2010 approval (Figure 1). Additionally, denosumab use was not associated with patient history of renal disease (Supplementary Table S4), in line with prior studies.22,32 In the US setting, it may therefore be possible to increase the overall value of prostate cancer care by reserving denosumab for patients who are not good candidates for zoledronic acid – such as those with renal disease – or by lowering the price of denosumab to ranges where it is more likely to be cost effective.
This study has several limitations. Castration sensitive vs. resistant status is not available in claims. We inferred CRPC by initiation of cancer treatments. The timing of treatment initiation may or may not be proximate to the emergence of clinical CRPC, as some patients may prefer to defer CRPC treatment. The proportion of patients who received BMAs within 180 days after clinical CRPC is likely to be slightly lower than the proportion who did so within 180 days after CDT, which we measured as a proxy. Our claims-based ascertainment of bone metastases is likely to be imperfect and result in some misclassification. However, a similar algorithm was highly specific for bone metastases,33 and BMA utilization was robust to sensitivity analysis (Supplementary Table S1). Our sample size was insufficient to assess the role of endocrinology involvement in multivariable analysis. Finally, SEER-Medicare is limited to elderly patients residing in SEER regions and may not be generalizable to other populations.
Conclusions
The majority of patients with mCRPC and bone metastases receive guideline-concordant BMA therapy at SRE-prevention doses. However, these patients commonly initiate BMA therapy at SRE-prevention doses prior to development of CRPC, when BMA therapy may not yet be warranted. Among patients with mCRPC and no bone metastasis, there may be a greater proportion of patients who would benefit from BMA therapy for osteoporotic fracture prevention.
Supplementary Material
Funding
This work was supported by the National Cancer Institute at the National Institutes of Health [grant numbers 1R03CA259863-01 to A.P.M, and P30CA008748].
Role of the Funder
The funding source had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.
Author Disclosures
APM and AS have no potential conflicts of interest to disclose.
AMM declares stock ownership in Johnson & Johnson, DNA, and Teladoc Health.
KP declares stock ownership in T2 Biosystems, Adicet Bio, Inc., Chinook Therapeutics, Codexis, and Vincerx Pharma.
AF declares consulting for Amgen.
MM declares consulting and/or advisory arrangements with Bayer, Novartis, Advanced Accelerator Applications, ORIC Pharmaceuticals, Johnson & Johnson, Curium Pharma, Athenex, Exelixis, AstraZeneca, and Amgen, paid travel for Endocyte and Fujifilm, and research funding from Bayer, Sanofi, Endocyte, Progenics, Corcept Therapeutics, Roche/Genentech, and Janssen.
Footnotes
Prior Presentation: American Society of Clinical Oncology Genitourinary Symposium, 2022
Data Sharing:
The data underlying this article were provided by the Centers for Medicare and Medicaid Services (CMS) under license. Data are available from CMS on request and establishment of data use agreement.
References
- 1.Oster G, Lamerato L, Glass AG, et al. Natural history of skeletal-related events in patients with breast, lung, or prostate cancer and metastases to bone: a 15-year study in two large US health systems. Support Care Cancer Off J Multinatl Assoc Support Care Cancer. 2013;21(12):3279–3286. doi: 10.1007/s00520-013-1887-3 [DOI] [PubMed] [Google Scholar]
- 2.Baek YH, Jeon HL, Oh IS, Yang H, Park J, Shin JY. Incidence of skeletal-related events in patients with breast or prostate cancer-induced bone metastasis or multiple myeloma: A 12-year longitudinal nationwide healthcare database study. Cancer Epidemiol. 2019;61:104–110. doi: 10.1016/j.canep.2019.05.013 [DOI] [PubMed] [Google Scholar]
- 3.Broder MS, Gutierrez B, Cherepanov D, Linhares Y. Burden of skeletal-related events in prostate cancer: unmet need in pain improvement. Support Care Cancer Off J Multinatl Assoc Support Care Cancer. 2015;23(1):237–247. doi: 10.1007/s00520-014-2437-3 [DOI] [PubMed] [Google Scholar]
- 4.NCCN Clinical Practice Guidelines in Oncology: Prostate Cancer, Version 4.2019. Published online August 19, 2019. Accessed January 17, 2020. https://www.nccn.org/professionals/physician_gls/pdf/prostate.pdf
- 5.Gillessen S, Attard G, Beer TM, et al. Management of Patients with Advanced Prostate Cancer: The Report of the Advanced Prostate Cancer Consensus Conference APCCC 2017. Eur Urol. 2018;73(2):178–211. doi: 10.1016/j.eururo.2017.06.002 [DOI] [PubMed] [Google Scholar]
- 6.Camacho PM, Petak SM, Binkley N, et al. AMERICAN ASSOCIATION OF CLINICAL ENDOCRINOLOGISTS/AMERICAN COLLEGE OF ENDOCRINOLOGY CLINICAL PRACTICE GUIDELINES FOR THE DIAGNOSIS AND TREATMENT OF POSTMENOPAUSAL OSTEOPOROSIS-2020 UPDATE. Endocr Pract Off J Am Coll Endocrinol Am Assoc Clin Endocrinol. 2020;26(Suppl 1):1–46. doi: 10.4158/GL-2020-0524SUPPL [DOI] [PubMed] [Google Scholar]
- 7.Wright NC, Looker AC, Saag KG, et al. The recent prevalence of osteoporosis and low bone mass in the United States based on bone mineral density at the femoral neck or lumbar spine. J Bone Miner Res Off J Am Soc Bone Miner Res. 2014;29(11):2520–2526. doi: 10.1002/jbmr.2269 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Shahinian VB, Kuo YF, Freeman JL, Goodwin JS. Risk of Fracture after Androgen Deprivation for Prostate Cancer. N Engl J Med. 2005;352(2):154–164. doi: 10.1056/NEJMoa041943 [DOI] [PubMed] [Google Scholar]
- 9.Giri S, Zhu W, Wang R, et al. Underutilization of guideline‐recommended supportive care among older adults with multiple myeloma in the United States. Cancer. 2019;125(22):4084–4095. doi: 10.1002/cncr.32428 [DOI] [PubMed] [Google Scholar]
- 10.Zhou J, Sweiss K, Patel PR, et al. Racial differences in intravenous bisphosphonate utilization among older patients with multiple myeloma. J Clin Oncol. 2019;37(15_suppl):e18165–e18165. doi: 10.1200/JCO.2019.37.15_suppl.e18165 [DOI] [Google Scholar]
- 11.Lepisto EM, Tangirala K, Vandergrift JL, et al. Concordance to National Comprehensive Cancer Network (NCCN) clinical practice guideline (GL) drug therapy recommendations for metastatic breast cancer: An analysis from the commercial managed care claims database PharMetrics (PM). J Clin Oncol. 2010;28(15_suppl):1032–1032. doi: 10.1200/jco.2010.28.15_suppl.1032 [DOI] [Google Scholar]
- 12.Liede A, Wade S, Lethen J, et al. An Observational Study of Concomitant Use of Emerging Therapies and Denosumab or Zoledronic Acid in Prostate Cancer. Clin Ther. 2018;40(4):536–549.e3. doi: 10.1016/j.clinthera.2017.12.015 [DOI] [PubMed] [Google Scholar]
- 13.Hagiwara M, Delea TE, Cong Z, Chung K. Utilization of intravenous bisphosphonates in patients with bone metastases secondary to breast, lung, or prostate cancer. Support Care Cancer Off J Multinatl Assoc Support Care Cancer. 2014;22(1):103–113. doi: 10.1007/s00520-013-1951-z [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Oster G, Lamerato L, Glass AG, et al. Use of intravenous bisphosphonates in patients with breast, lung, or prostate cancer and metastases to bone: a 15-year study in two large US health systems. Support Care Cancer Off J Multinatl Assoc Support Care Cancer. 2014;22(5):1363–1373. doi: 10.1007/s00520-013-2094-y [DOI] [PubMed] [Google Scholar]
- 15.Fizazi K, Carducci M, Smith M, et al. Denosumab versus zoledronic acid for treatment of bone metastases in men with castration-resistant prostate cancer: a randomised, double-blind study. Lancet Lond Engl. 2011;377(9768):813–822. doi: 10.1016/S0140-6736(10)62344-6 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Lipton A, Fizazi K, Stopeck AT, et al. Superiority of denosumab to zoledronic acid for prevention of skeletal-related events: a combined analysis of 3 pivotal, randomised, phase 3 trials. Eur J Cancer Oxf Engl 1990. 2012;48(16):3082–3092. doi: 10.1016/j.ejca.2012.08.002 [DOI] [PubMed] [Google Scholar]
- 17.Henry DH, Costa L, Goldwasser F, et al. Randomized, double-blind study of denosumab versus zoledronic acid in the treatment of bone metastases in patients with advanced cancer (excluding breast and prostate cancer) or multiple myeloma. J Clin Oncol Off J Am Soc Clin Oncol. 2011;29(9):1125–1132. doi: 10.1200/JCO.2010.31.3304 [DOI] [PubMed] [Google Scholar]
- 18.Autio KA, Farooki A, Glezerman IG, et al. Severe Hypocalcemia Associated With Denosumab in Metastatic Castration-Resistant Prostate Cancer: Risk Factors and Precautions for Treating Physicians. Clin Genitourin Cancer. 2015;13(4):e305–e309. doi: 10.1016/j.clgc.2014.11.008 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Reed SD, Radeva JI, Glendenning GA, Saad F, Schulman KA. Cost-effectiveness of zoledronic acid for the prevention of skeletal complications in patients with prostate cancer. J Urol. 2004;171(4):1537–1542. doi: 10.1097/01.ju.0000116777.94426.60 [DOI] [PubMed] [Google Scholar]
- 20.Snedecor SJ, Carter JA, Kaura S, Botteman MF. Denosumab versus zoledronic acid for treatment of bone metastases in men with castration-resistant prostate cancer: a cost-effectiveness analysis. J Med Econ. 2013;16(1):19–29. doi: 10.3111/13696998.2012.719054 [DOI] [PubMed] [Google Scholar]
- 21.Kanellias N, Gavriatopoulou M, Terpos E, Dimopoulos MA. Management of multiple myeloma bone disease: impact of treatment on renal function. Expert Rev Hematol. 2018;11(11):881–888. doi: 10.1080/17474086.2018.1531702 [DOI] [PubMed] [Google Scholar]
- 22.Mitchell AP, Mishra A, Panageas KS, Lipitz-Snyderman A, Bach PB, Morris MJ. Real-World Use of Bone Modifying Agents in Metastatic Castration-Sensitive Prostate Cancer. J Natl Cancer Inst. Published online October 1, 2021:djab196. doi: 10.1093/jnci/djab196 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Klabunde CN, Potosky AL, Legler JM, Warren JL. Development of a comorbidity index using physician claims data. J Clin Epidemiol. 2000;53(12):1258–1267. [DOI] [PubMed] [Google Scholar]
- 24.Jain R, Bach P. Hospital Outpatient versus Physician Office Cost for Physician Administered Cancer Drugs. Memorial Sloan-Kettering Cancer Center; 2017. Accessed October 14, 2021. https://www.drugpricinglab.org/wp-content/uploads/2017/01/Hospital-outpatient-versus-doctor-office-cost-for-physician-administered-cancer-drugs.pdf [Google Scholar]
- 25.Fitch K, Pyenson B. Site of Care Cost Differenes for Medicare Patients Receiving Chemotherapy. Milliman; 2011. Accessed April 23, 2018. http://us.milliman.com/uploadedFiles/insight/health-published/site-of-service-cost-differences.pdf [Google Scholar]
- 26.Lipitz-Snyderman A, Atoria CL, Schleicher SM, Bach PB, Panageas KS. Practice Patterns for Older Adult Patients With Advanced Cancer: Physician Office Versus Hospital Outpatient Setting. J Oncol Pract. 2019;15(1):e30–e38. doi: 10.1200/JOP.18.00315 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Mark M, Thürlimann B, Ribi K, et al. Patterns of care for patients with metastatic bone disease in solid tumors: A cross-sectional study from Switzerland (SAKK 95/16). J Bone Oncol. 2020;21:100273. doi: 10.1016/j.jbo.2019.100273 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Stoffel ST, von Moos R, Thürlimann B, et al. Patterns of care and economic consequences of using bone-targeted agents for castration-sensitive prostate cancer patients with bone metastases to prevent skeletal-related events in Switzerland - the SAKK 95/16 prostate study. Swiss Med Wkly. 2021;151:w20464. doi: 10.4414/smw.2021.20464 [DOI] [PubMed] [Google Scholar]
- 29.Stopeck A, Rader M, Henry D, et al. Cost-effectiveness of denosumab vs zoledronic acid for prevention of skeletal-related events in patients with solid tumors and bone metastases in the United States. J Med Econ. 2012;15(4):712–723. doi: 10.3111/13696998.2012.675380 [DOI] [PubMed] [Google Scholar]
- 30.Carter JA, Botteman MF. Health-economic review of zoledronic acid for the management of skeletal-related events in bone-metastatic prostate cancer. Expert Rev Pharmacoecon Outcomes Res. 2012;12(4):425–437. doi: 10.1586/erp.12.31 [DOI] [PubMed] [Google Scholar]
- 31.Lothgren M, Bracco A, Lucius B, et al. PCN116 Cost-Effectiveness of Denosumab Versus Zoledronic Acid (ZA) for the Prevention of Skeletal-related Events (SRE) in Patients with Bone Metastases from Solid Tumors in the Netherlands. Value Health. 2011;14(7):A455. doi: 10.1016/j.jval.2011.08.1217 [DOI] [Google Scholar]
- 32.Gupta A, Wang P, Ali SA, et al. Use of Bone-Modifying Agents Among Medicare Beneficiaries With Multiple Myeloma. JAMA Oncol. Published online December 12, 2019. doi: 10.1001/jamaoncol.2019.5426 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Sathiakumar N, Delzell E, Yun H, et al. Accuracy of Medicare Claim–based Algorithm to Detect Breast, Prostate, or Lung Cancer Bone Metastases: Med Care. 2017;55(12):e144–e149. doi: 10.1097/MLR.0000000000000539 [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
The data underlying this article were provided by the Centers for Medicare and Medicaid Services (CMS) under license. Data are available from CMS on request and establishment of data use agreement.