Population-based risk factors and resource utilization for HCC: US perspective

OBJECTIVE

Hepatocellular carcinoma (HCC) is a deadly cancer with limited treatment options. HCC cases in the United States (US) were identified from a claims database to analyze the risk factors, the health care provider referral patterns, and treatment options in actual (real-world) clinical settings.

METHODS

MarketScan, a health care claims database from Thomas Reuters covering 18 million lives yearly and all US census regions from 2002 to 2008, was used to identify HCC patients and obtain data on patient characteristics, health care providers, and treatment utilization (i.e., medications, interventions).

RESULTS

HCC cases (n = 4406) were identified with an annual incidence of 0.4 per 1000 covered lives (i.e., those currently enrolled in a health care plan) from 2002 to 2008. Nonalcoholic fatty liver disease (NAFLD)/nonalcoholic steatohepatitis (NASH) was the most common underlying etiologic risk factor (59%), followed by diabetes (36%) and hepatitis C virus infection (22%). Primary care/internal medicine providers managed the majority of cases (55%); a minority were seen by an oncologist (24%). Only 22% of cases known to have cirrhosis were undergoing HCC screening prior to diagnosis. Type of provider did not change significantly after the diagnosis was made. Systemic chemotherapy was the most commonly used treatment (32.8%); however, only 6% received sorafenib, the only approved drug for HCC. Limitations include lack of patient records and potential for physician coding variances.

CONCLUSION

The incidence of HCC in the database was 0.4 per 1000 persons. NAFLD/NASH and type 2 diabetes mellitus, along with hepatitis C virus infection, were the major etiologic risk factors associated with HCC. This claims database analysis suggests a gap exists between screening and treatment guidelines and practice patterns, implying a need for greater health care provider awareness and education.

Medical costs of sorafenib compared with sunitinib in treatment of patients <65 years with renal cell carcinoma: A retrospective claims database analysis

e17536

Background: Sorafenib (SR) and sunitinib (SU) are FDA approved (12/05 and 1/06, respectively) tyrosine kinase inhibitors for patients (pts) with advanced renal cell carcinoma (RCC). Little is known of their cost impact. We conducted this study to quantify overall direct costs (inpatient, outpatient, pharmacy) for each treatment in privately insured pts (<65 y). Given differences between private and public insurer payment scales, pts ≥65 y were not included in this study but are the subject of a planned further analysis. Methods: A retrospective US claims-based study was conducted using data covering all US census regions for ≥18 million lives from MarketScan MedStat (1/02–12/07). Inclusion criteria were ≥2 RCC claims (ICD-9 189.0, 198.0), continuous health care coverage, >180 days of coverage before RCC diagnosis. SR and SU pts were identified based on oral therapy after initial RCC-related claim (intent-to-treat). Observation period was from first drug-dispensing date until ≤12 mo or first of therapy switch, nephrectomy, disenrollment, or study end (12/31/07). Univariate and multivariate Tobit analyses were conducted; control factors included age, sex, region, plan type, comorbidity, prior Tx/procedures, and time since RCC Dx. Results: Of 10,462 RCC pts identified, 144 and 220 received initial therapy with SR and SU, respectively. In the 180 days before RCC diagnosis, total direct medical costs, baseline demographics, and comorbidities were similar between groups. The univariate incremental total monthly medical cost for SU was $2,049 (P<.001) more than for SR, representing yearly costs for SU $24,588 more than for SR. Multivariate analyses for incremental total monthly costs for SU also remained significant at $1,399 (P<.001). Conclusions: Retrospective analysis of this US claims database for RCC pts <65 treated first with SR showed statistically significant lower total medical costs (particularly inpatient costs) than for pts treated first with SU.

[Table: see text]

[Table: see text]

Pattern of utilization of pegfilgrastim in patients with chemotherapy-induced neutropenia: A retrospective analysis of administrative claims data

9624

Background: Pegfilgrastim is a long-acting granulocyte colony-stimulating factor (G-CSF) used to prevent or treat febrile neutropenia associated with myelosuppressive anticancer therapies. According to the prescribing information, pegfilgrastim should not be administered within 14 days before or 24 hours after cytotoxic chemotherapy because of the potential for myeloid toxicity. This study examined use patterns of pegfilgrastim in real-life practice. Methods: Analysis of health insurance claims data in 2000- 2007 from > 35 large health plans across the US was conducted. Patients who had a cancer diagnosis and chemotherapy within 120 days of their first pegfilgrastim injection were identified. The proportion of pegfilgrastim injections that were followed by administration of chemotherapy within 11 and 9 days was calculated. Analysis was also stratified by cancer type [Non-Hodgkin's lymphoma (NHL), lung, breast]. Results: A total of 13,526 cancer patients received 57,118 pegfilgrastim injections. NHL, lung, and breast cohorts comprised 2,722, 2,772, and 4,955 patients, respectively. Mean age (SD) was 55.0 (11.6) and women represented 65.9% of study population. Among all cancer types, 19.2% of pegfilgrastim injections had a chemotherapy claim within the following 11 days. This pattern of use was the highest in NHL (18.9%), followed by lung (17.1%), and breast (16.2%). Similar results were observed in the 9-day sensitivity analysis (see Table ). Conclusions: Based on the retrospective analysis of this administrative claims database, the use of pegfilgrastim within 11 days of an administration of chemotherapy was observed in 15–20% of cases which is inconsistent with the recommended guidelines. Pegfilgrastim use in these situations may have the potential to increase sensitivity of rapidly dividing myeloid cells to cytotoxic chemotherapy. Further research is being conducted to assess the related clinical and economic impact of this pattern of usage. [Table: see text]

[Table: see text]

Dosing patterns in patients with renal cell carcinoma treated with sorafenib or sunitinib: A retrospective claims database analysis

5097

Background: Tyrosine kinase inhibitors (TKIs; sorafenib [SR], sunitinib [SU]) are FDA approved for the treatment of advanced renal cell carcinoma (RCC). Dose reduction may be needed for management of treatment-related adverse events. We analyzed the dose reduction patterns in pts with RCC treated with SR, approved Dec 05, or SU, approved Jan 06. Methods: A retrospective analysis was conducted using data from a US claims-based database (MarketScan MedStat) covering ≥18 million lives for 2002–2008 in all US census regions. Inclusion criteria were ≥2 claims for RCC (ICD9 189.0 or 198.0), continuous healthcare coverage, >180 days of coverage before RCC diagnosis, no claim for SR or SU prior to RCC diagnosis, initial standard daily RCC dose per package insert recommendation (800 mg for SR or 50 mg for SU), and ≥2 consecutive dispensings. Initial episode was defined as the time from the date of the first drug dispensing (index date) to the first switch to another TKI, end of healthcare coverage, end of treatment, or March 31, 2008. Pt and pt-time level analyses for dose reductions between treatments were conducted. Results: SR- (n = 189) and SU-treated (n = 304) pts met the inclusion criteria. No significant differences in baseline demographics existed between the groups except for a higher incidence of stroke (7.9% vs 3.6%, p = 0.037) and other cancer site (93.7% vs 87.8%, p = 0.036) in the SR group. Significantly more pts who received SU (23%) required dose reductions during the first 3 months of therapy and for all initial episodes than did pts who received (4.2%) SR (p < 0.001 for both). For all episodes, the mean time to dose reduction was 104 days for SU versus 162 days with SR (p = 0.0028). The majority (65%) of dose reductions with SU occurred within the first 3 months, whereas dose reduction with SR was 25% during this time (p < 0.001). After controlling for different lengths of exposure time, 2 to 6 times more dose reductions were observed in pts who received SU than with SR (p < 0.001). Conclusions: This US retrospective claims analysis showed more dose reductions, including number of patients and total number of days, in pts who initially received sunitinib compared with those who received sorafenib.

[Table: see text]

Retrospective claims database analysis of elderly compared with nonelderly patients (pts) with newly diagnosed hepatocellular carcinoma (HCC)

9552

Background: HCC is the most common type of liver cancer in the US and its incidence has been rising. We assessed whether differences existed in risk factors and treatment practices in pts with HCC by age. Methods: De-identified, individual-level healthcare claims data from a database (MarketScan MedStat) covering all US census regions for ≥18 million lives in 2002–2008 were retrospectively analyzed. Pts had ≥2 claims for primary HCC (ICD9 155.0), continuous healthcare coverage, >180 days of coverage before HCC, and no prior claims for secondary liver cancer (ICD9 197.7). Pts were followed longitudinally from HCC diagnosis until end of healthcare coverage or 3/31/08. Pts were stratified by age (<65 or ≥65 y), and treatment was analyzed using prevalence and time to initiation. Results: Of 4,153 pts with HCC identified, 46.3% were ≥65 y. Risk factors in the groups were nonalcoholic fatty liver disease/nonalcoholic steatohepatitis (NAFLD/NASH, 59% for <65 vs 50% for ≥65), hepatitis C (32% vs 11%), diabetes (29% vs 40%), alcoholic cirrhosis (16% vs 7%), and hepatitis B (9% vs 3%). Treatments were similar between groups; however, more pts in the younger group received transplantation, IV chemotherapy, and hospice care ( Table ). Twice as many older pts did not receive any of the considered treatment compared with younger pts. Sorafenib, approved for unresectable HCC in 11/07, was used in a limited number of pts in the groups. Most noncurative treatment options were initiated sooner in pts ≥65 y than those <65 y, whereas the opposite holds for the 2 main curative treatments (transplantation and cancer removal surgery). Conclusions: Regardless of age, principal risk factors associated with HCC were NAFLD/NASH, diabetes, and hepatitis C. Based on this retrospective claims analysis, variations between treatment options and time to initiation did differ between the groups, suggesting that underlying comorbidities and practice patterns affect treatment options in elderly and nonelderly pts with HCC.

[Table: see text]

[Table: see text]

Budget impact analysis of sargramostim use in patients with chemotherapy-induced neutropenia

e20596

Background: Myeloid growth factors are used to treat and prevent chemotherapy-induced neutropenia (CIN). Filgrastim and its long-acting version pegfilgrastim are granulocyte colony-stimulating factors (G-CSF), whereas sargramostim is a dual granulocyte- macrophage colony-stimulating factor (GM-CSF). This study analyzed the budget impact of substituting GM-CSF for G-CSF in the management of CIN from the perspective of a US health plan. Methods: A spreadsheet model was developed to compute annual and per-member-per-month (PMPM) costs associated with CSFs. Inputs included cancer prevalence, the proportion of patients receiving chemotherapy and G/GM-CSFs, incidence and cost of relevant adverse events (e.g., bone pain), and G/GM-CSF drug acquisition and administration costs. Incidence and cost of infection- and febrile neutropenia-related hospitalizations, based on recent analysis of medical insurance claims data, were also used. Cost savings (2006 USD) were assessed for utilization share switches from G-CSF to GM-CSF. Results: For a health plan with 1 million members, an estimated 976 patients received G/GM-CSF annually. Increasing baseline utilization shares for pegfilgrastim, filgrastim, and sargramostim of 70/30/0%, respectively, to alternative shares of 50/25/25% yielded substantial cost savings (see Table ), primarily related to G/GM-CSF acquisition and administration costs. Savings for patients switching from pegfilgrastim were greater than for patients switching from filgrastim. Results were sensitive to assumptions for drug cost and frequency of administration, but cost savings were observed for most scenarios. Conclusions: This study suggests that health plans can realize substantial cost savings by substituting sargramostim for filgrastim and pegfilgrastim in CIN patients. With 25% of sargramostim substitution, the cost saving could reach ≈$2 million for a health plan with 1 million members, or a saving of 16 cents per member per month.

[Table: see text]

[Table: see text]

Cost utility analysis of early adjuvant letrozole or anastrozole versus tamoxifen in postmenopausal women with early invasive breast cancer: the UK perspective

Five years with the aromatase inhibitors letrozole or anastrozole is clinically superior to 5 years tamoxifen in postmenopausal women with early breast cancer. This paper analyses the cost-effectiveness of the aromatase inhibitors compared to tamoxifen using the same health economic model. A Markov model describes lifetime incidence of breast cancer events and treatment-related adverse events. Probabilities of disease progression, adverse events, and utility values were estimated using secondary sources; costs of breast-cancer care were obtained from a primary costing study. The incremental cost per QALY gained of letrozole vs. tamoxifen is 10,379pounds (95% CI 6,705-23,574pounds), and of anastrozole versus tamoxifen is 11,428pounds (95% CI 6,211-48,795pounds). If a 5-year carry over effect for the reduction in breast cancer events is assumed, the incremental costs per QALY gained compared to tamoxifen are 6,253pounds (95% CI 3,675-14,766pounds) for letrozole and 7,015pounds (95% CI 3,316-31,997pounds) for anastrozole. Five years of letrozole or anastrozole therapy is cost-effective in postmenopausal women with early breast cancer. Though the respective confidence intervals show significant overlap, letrozole has a 95% probability of being more cost-effective than tamoxifen at a 20,000pounds QALY value, whilst anastrozole has an 85% probability.

Renal function among cancer patients with bone metastases treated with zoledronic acid in a real world setting

19540

Background: Zoledronic acid (ZA) is a bisphosphonate that reduces skeletal related events (SREs) in cancer patients (pts) with bone metastases but can impact renal function in some pts. This retrospective study examined serum creatinine (sCr) elevations and subsequent treatment changes among pts receiving ZA in community based medical practices. Methods: Data were obtained from the Varian Medical Oncology database of electronic medical records from 17 US oncology practices. Cancer pts with bone metastases and at least one ZA infusion in 2002–06 were followed; all available sCr levels were examined. SCr measurements ≤4 weeks after a ZA infusion were considered levels while on ZA. Elevated sCr was defined as an increase over baseline of ≥0.5 mg/dL or a doubling from baseline for pts with baseline levels <1.4 mg/dL (or ≥1.0 mg/dL for higher baseline levels). Pts with a sCr elevation on ZA were followed to observe changes in ZA dosing and sCr. Results: The study included 875 pts (318 with breast cancer, 131 lung, 154 prostate, 23 multiple myeloma, and 249 other or unknown cancer) with ≥1 sCr level during baseline and ≥1 (median 6) while on ZA. Median age was 66, 41% were male, and 90% had baseline sCr under 1.4 mg/dL. Pts received a median of 6 ZA infusions. SCr elevations occurred in 87 pts (10%), at a median of 19 weeks (range 0.6–126) after the start of ZA. Pts with baseline sCr over 2.0 mg/dL were more likely to experience an elevation (11 pts, 33%). Following the elevated sCr, 37 pts (43%) discontinued ZA; dose reductions and delays were infrequent. Among 49 pts who remained on ZA after elevated sCr and had further sCr while on ZA, 20 (41%) returned to within 10% of baseline, at a median of 7 weeks (range 0.3–31) after the first measured elevation. The proportion returning to normal is a conservative estimate given limited follow-up sCr data. Conclusions: The observed incidence of elevated sCr of 10% in this community based study is similar to clinical trials. Most pts either discontinued ZA or returned to baseline levels despite continued ZA treatment. Few practitioners adhered to recommendations to withhold ZA until levels return to within 10% of baseline.

No significant financial relationships to disclose.

Solid tumor cancer patients treated with zoledronic acid experienced reduced skeletal related events as compared to untreated patients in clinical practice claims database

6596

Background: Patients (pts) with bone metastasis are at high risk for skeletal related events (SREs), leading to increased morbidity, mortality, and decreased QOL. Methods: To study the impact of zoledronic acid on SRE rates, follow-up duration (from 1st bone metastasis diagnosis to last claim) in treated pts as compared to the experience of non bisphosphonate-treated pts (untreated), and to investigate relative impact of zoledronic acid (ZA) frequency of administration using data from PharMetrics integrated claims database of 80 health plans across the U.S, from Jan 02 to Oct. 05. Results: There were 1,518 ZA treated-pts and 3308 untreated-pts, who met study inclusion criteria of having breast, prostate, or lung cancers, with confirmed bone metastasis and SRE experience. At baseline, breast cancer pts (N=1,799) had lowest Charlson Co-morbidity Index (CCI); lung cancer pts (N=2,413) had highest CCI (p<0.05). Prostate cancer pts (N=622) were significantly older (p<0.05). The average monthly ZA prescription rate (Rx) was 0.73 with average total Rxs of 7.82 over the follow-up duration. ZA use reduced monthly SRE rate by 30% from 0.41 ± 0.4 in untreated pts to 0.29 ± 0.3 in treated pts (p<0.001) and prolonged follow-up durations in months from 9.4 ± 9.8 in untreated to 12.2 ± 9.0 in treated pts (p<0.001). Regression models of SRE rate or follow-up duration as the dependent variable investigating the impact of ZA prescribing patterns indicated that pts who followed ZA label dose of 4 mg every q3, q4 experienced the lowest SRE rate (p<0.001) and had longer follow-up duration (p<0.005) as compared untreated pts. ZA use almost doubled time to 1st SRE from 101±165 days in untreated pts, to 185±210 days in treated pts (p<0.0001). ZA also delayed time between 1st to 2nd SRE from 85 days in untreated to 111 days in treated pts (p<0.05). Conclusion: Zoledronic acid significantly reduced SRE rate in cancer pts with bone metastasis by 30%, increased follow-up duration by almost 3 months, doubled time to 1st SRE, delayed time to 2nd SRE, with its label dose of 4 mg every 3 to 4 weeks achieving the best outcome.

No significant financial relationships to disclose.

Deferasirox for myelodysplastic syndrome patients with transfusional iron overload: A budget impact analysis

17019

Background: Myelodysplastic syndromes (MDS) affect 12–20,000 in the US yearly. Blood transfusions, a mainstay of treatment, place MDS patients (pts) at risk for complications of transfusional iron overload (IO). Although recent data suggests iron accumulation significantly reduces survival in low and intermediate -1 risk MDS, incremental cost of adding chelators to health plan formularies are unknown. The economic impact of treating transfusional IO in MDS with two commercially available chelating agents were estimated from a health care plan (hcp) perspective. Methods: An Excel based model was developed to ascertain the incremental cost per treated MDS pt associated with oral deferasirox (DSX) relative to subcutaneous (sq) deferoxamine (DFO). Annual Per Member Per Month (PMPM) cost was calculated by dividing differences in annual expenditures for DSX and DFO by total plan membership. Data from published literature and pivotal clinical trials were used to estimate MDS prevalence and chelation eligibility. Costs associated with infusing sq DFO were based on medical claims analysis from 2001 to 2004. Medication costs were based on wholesale acquisition costs assuming 100% compliance. Sensitivity analyses were performed by varying model parameters. Results: A hcp of one million pts is estimated to have 119 pts with MDS, of which 22% are eligible for iron chelation therapy. The base-case model resulted in an incremental annual saving of $389.48 per treated MDS pt and an incremental budget impact of $0.0 PMPM of the hcp (function of relatively few MDS pts spread over million in hcp) associated with the use of DSX. Overall, 34.2% of DSX cost was offset by the absence of infusion costs, the key cost driver. Sensitivity analyses of varying pt weight and infusion costs showed the model results were robust. Conclusions: The annual budget impact of switching MDS patients with IO to DSX from DFO will potentially result in cost-savings for US health plans. These results should be considered in the selection of chelation therapy.

No significant financial relationships to disclose.

Relationship between compliance with imatinib mesylate and medical costs for patients with CML and GIST

6618

Background: Imatinib mesylate, an oral tyrosine kinase inhibitor for treatment of chronic myeloid leukemia (CML) and gastrointestinal stromal tumors (GIST), has been found both effective and cost-effective. This study examines the relationship between imatinib compliance and health care costs for patients (pts) with CML and GIST. Methods: This retrospective study used claims data from a large national US health plan. Pts had =2 imatinib claims from 6/1/01–3/31/05; =1-year follow-up after the 1st imatinib claim; and diagnosis of CML (ICD9 205.1x) or GIST (ICD9 159.0, 159.8, 159.9 or evaluation of claims and attribution of GIST by independent hematologist/oncologist). Compliance was measured with a medication possession ratio ((days of imatinib during follow-up / days of follow- up)*100). Compliance categories were good, =90%; medium, 70–89.9%; poor, <70%. Total follow-up medical and health care (medical + pharmacy) costs were measured and compared across compliance categories; t-tests compared mean costs. Results: 374 CML and 91 GIST pts had mean age of 52.1 ± 14.0 years. 274 (59%) pts were male. 234 (50%) of pts had 12–24 months follow-up; 231 (50%) had 25–57 months. Mean and median compliance across all pts were 69.4% and 79.7%, respectively. 122 (33%) CML and 32 (35%) GIST pts had good compliance; 149 (40%) CML and 42 (46%) GIST pts had poor compliance. Mean medical costs across all pts were lower and less variable with good compliance ($22,882 ± 22,791) than with medium ($40,366 ± 68,186), p=0.007) and poor ($104,961 ± 190,559), p<0.001). Similarly, mean medical costs for CML pts with good, medium, and poor compliance were $21,456 (±22,127), $41,537 (±72,238, p=0.008), and $117,004 (±211,292, p<0.001), respectively. Mean medical costs for GIST pts were: good=$28,318 (±24,781); medium=$33,270 (±35,356, p=0.584); and poor=$62,235 (±68,751, p<0.001). Mean total health care costs were 21.7% lower for pts with good compliance relative to those with poor. Conclusions: Good imatinib compliance was associated with significantly lower medical costs. Mean total medical costs were 78% lower with good compliance relative to poor. Compliance is an important treatment issue for both clinical and medical cost outcomes.

No significant financial relationships to disclose.

Compliance and persistency with a new oral iron chelator in patients with myelodysplastic syndromes

17510

Background: Patients with myelodysplastic syndromes (MDS) frequently receive blood transfusions and are at risk for iron overload (IO). The aim of this study was to estimate compliance and persistency rates of a new oral iron chelator, deferasirox, in patients with MDS in a real world setting. Methods: A retrospective analysis of patients provided deferasirox by Accredo Nova Factor database was conducted. Patients diagnosed with MDS and receiving deferasirox were followed from Jan 1, 2006 to Nov 30, 2006. Patients with at least 2 filled prescriptions were included in the analysis. Compliance was defined by medication possession ratio (MPR = # of days of medication supplied/# of days in refill interval). MPR>0.80 is categorized as good compliance. Persistency was defined as the continuous prescription refill without an interruption of longer than 30 days and Kaplan Meier curves were analyzed. Reasons for discontinuation and differences in compliance rates were explored. Results: A total of 294 patients with MDS and at least 2 filled prescriptions filled were included in the analysis. The average age was 72.2 ±12.3 years and 53.7% were male. 64.1%, 30.3%, and 3.7% of patients were covered by private insurance, Medicare, and Medicaid, respectively. The average compliance with deferasirox was 0.89 ± 0.17. 67.9% of patients had an MPR greater than 0.90 (data left-skewed as ratio truncated at 1.0). Overall, 86.4% of patients were considered persistent with deferasirox during the analysis period. Reasons for discontinuing therapy included side effects (n=14; 5%), physician choice (n =10; 3%), patient choice (n=4; 1%), or a change in provider (n=6; 2%). Conclusion: MDS patients with transfusional IO when treated with deferasirox are highly compliant and persistent. Given the decreased survival associated with IO due to chronic blood transfusions in low and intermediate-1 risk MDS patients, compliance and persistency are important factors to consider when selecting therapies.

[Table: see text]

Cost effectiveness of bisphosphonates in the management of breast cancer patients with bone metastases

BACKGROUND

Bisphosphonates are recommended to prevent skeletal related events (SREs) in patients with breast cancer and bone metastases (BCBM). However, their clinical and economic profiles vary from one agent to the other.

MATERIALS AND METHODS

Using modeling techniques, we simulated from the perspective of the UK's National Health Service (NHS) the cost and quality adjusted survival (QALY) associated with five commonly-used bisphosphonates or no therapy in this patient population. The simulation followed patients into several health states (i.e. alive or dead, experiencing an SRE or no SRE, and receiving first or second line therapy). Drugs costs, infusion costs, SREs costs, and utility values were estimated from published sources. Utilities were applied to time with and without SREs to capture the impact on quality of life.

RESULTS

Compared to no therapy, all bisphosphonates are either cost saving or highly cost-effective (with a cost per QALY < or = 6126 pounds sterlings). Within this evaluation, zoledronic acid was more effective and less expensive than all other options.

CONCLUSIONS

Based on our model, the use of bisphosphonates in breast cancer patients with bone metastases should lead to improved patient outcomes and cost savings to the NHS and possibly other similar entities.

Cost-effectiveness of zoledronic acid vs. pamidronate in the management of hormone refractory prostate cancer (HRPC) patients with bone metastases

14660

Background: Canadian guidelines recommend zoledronic acid (ZA) (4 mg every 3 weeks) in patients with hormone HRPC and asymptomatic or minimally symptomatic bone metastases to reduce skeletal-related events (SRE). However, IV pamidronate (90 mg every 3 weeks) (PA) is also routinely used in this setting in spite of no significant improvement in occurrence of SRE or pain compared to NT. Objectives: To assess the cost effectiveness of ZA, PA, or no bisphosphonate therapy (NT) in the management of prostate cancer patients with bone metastases in Canada. Methods: A literature-based decision analytic model was developed to estimate the incremental cost and quality-adjusted life years (QALY) associated with the 3 treatment options. The model included assumptions about SREs, mortality, drug and administration costs, cost of SREs, reduced quality of life due to SRE and bone pain, and therapy duration. Sensitivity analyses considered several scenarios in which various assumptions were used regarding treatment efficacy and QALY gains due to pain relief and SRE prevention. All costs were expressed in Canadian dollars (2004). Results: The cumulative number of SREs over a patient’s remaining lifetime (1.9 years) was estimated at 2.69 for PA and NT patients and 1.76 for ZA. Total discounted costs were $11,918 for NT, $17,593 for PA, and $19,312 for ZA. Compared with patients receiving NT or PA, quality-adjusted survival increased by 0.094 (range depending on scenario considered: 0.072 to 0.106) QALY per patient for those on ZA compared to PA or NT. Compared to NT, ZA resulted in a cost per QALY gained of $78,366 (range: $50,717 to $101,831). Compared to PA, ZA resulted in a cost per QALY gained of $18,343 (range: $2917 to $23,835) per QALY gained. PA was more expensive than NT but did not improve patient-related outcomes. Conclusions: For HRPC patients with bone metastasis, zoledronic acid appears to be the only clinically valuable and economically acceptable option in Canada with a cost effectiveness likely better than previously reported.

No significant financial relationships to disclose.

Impact of zoledronic acid (Z) on pain in prostate cancer patients with bone metastases in a randomised placebo-control trial

4638

Background: Bone metastases occur in up to 80% of advanced prostate cancer patients and can results in painful Skeletal-Related Events (SREs). Z is the only biphosphonate approved for metastatic prostate cancer that decreases the occurrence of SREs which include pathologic fractures, severe bone pain requiring radiation therapy, spinal cord compression, surgery to bone and hypercalcemia of malignancy which may lead to daily activity limitations. We performed retrospective analyses on data from a placebo-controlled study of Z to investigate which items of the Brief Pain Inventory (BPI) were most sensitive to changes in patient’s pain due to treatment. Methods: Prostate cancer patients with bone metastases received Zoledronate 4 mg (n = 201) or placebo (n = 201). The BPI was administered at baseline (visit 2), and visits 3 (weeks 3–4), 4 (weeks 6–8), and then every second visit (every 6 weeks) until study end. Changes in BPI individual item scores and the Pain Severity and Pain Interference scale scores were compared between Z and placebo groups. Post-baseline missing data was replaced by Last Observation Carried Forward (LOCF). Results: Z was statistically significantly better than placebo for changes in BPI items ‘pain at its worst’, ‘pain at its least’, ‘pain on average’, and the BPI pain severity scale (p < 0.05) at visit 4 (weeks 6–8), and visit 6 (week 12). For ‘pain right now’ there were statistically significant differences at visit 4, but not visit 6. Z was also superior to placebo for ‘pain right now’ and the Pain Severity and Interference scales at visit 12 (week 30–32) (p < 0.05). Finally Z was superior to placebo for ‘interference with general activities’ at visits 12 (week 30–32), 16 (weeks 42–44), and LOCF (p < 0.05). Conclusion: Our findings suggest that Z significantly reduces bone pain and general activity limitations in prostate cancer patients with bone metastases. Several items in the BPI are more sensitive to Z treatment including ‘pain at its worst’, ‘pain at its least’, ‘pain on average’, and the BPI Pain Severity scale.

[Table: see text]

Cost-effectiveness of letrozole and anastrozole as adjuvant therapy for hormone receptor positive early breast cancer in postmenopausal women

10577

Background: The BIG 1–98 and ATAC studies demonstrated that, in postmenopausal women with hormone receptor positive (HR+) early breast cancer, 5 years of initial adjuvant therapy with the aromatase inhibitors (AIs) letrozole (LET) or anastrozole (ANA) is superior to tamoxifen (TAM). The cost-effectiveness TAM, LET, and ANA have not been previously evaluated using a consistent methodology. Methods: A Markov model was used to estimate the incremental cost per quality-adjusted life year (QALY) gained with initial adjuvant therapy with LET vs TAM, ANA vs TAM, and LET vs ANA in postmenopausal women with HR+ early stage breast cancer from the US healthcare system perspective. Probabilities of recurrence (including contralateral tumor) and adverse events (endometrial cancer, thromboembolism, fractures, hypercholesterolemia, MI, and stroke) for TAM were based primarily on published US population-based studies and trials of prophylactic TAM vs placebo. Corresponding probabilities for LET and ANA were calculated by multiplying probabilities for TAM by estimated relative risks of LET vs TAM and ANA vs TAM from the BIG 1–98 and ATAC trials respectively. Other probabilities, costs, and health-state utilities were obtained from published studies. Expected lifetime costs and QALYs were estimated for a cohort of HR+ postmenopausal women with early breast cancer, aged 61 years at therapy initiation and discounted at 3% annually. Probabilistic sensitivity analyses were conducted to assess precision of results. Results: Incremental cost per QALY gained for LET vs TAM is $33,536 (95% CI $20,409 to $70,566) and for ANA vs TAM is $38,967 (95% CI $23,826 to $81,904). Compared with ANA, LET is less costly ($9,647 vs $10,190) and gains more QALYs (0.29 vs 0.26), although differences in costs (95% CI -$1,669 to $671) and QALYs (95% CI -0.16 to 0.22) are not statistically significant. Conclusions: In postmenopausal women with HR+ early breast cancer, adjuvant therapy with either LET or ANA is cost-effective from a US healthcare system perspective. Although LET dominates ANA in our base-case analysis, definitive conclusions regarding the cost-effectiveness of LET vs ANA must await results of comparative clinical studies.

[Table: see text]

Effect of zoledronic acid (Z) on pain in prostate cancer patients with bone metastases based on performance status

14544

Background: Z has demonstrated efficacy in reducing, preventing, delaying the occurrence of an SRE (Skeletal Related Event) including pathologic fractures, severe bone pain requiring radiation therapy, spinal cord compression, surgery to bone and hypercalcemia of malignancy which could lead to a loss of independence. Effects of Z on pain were analyzed in a subgroup of patients based on performance status. Methods: Subjects were randomised to Z 4mg (n = 201) or placebo (n = 201). The short form of the Brief Pain Inventory (BPI) was administered at baseline (visit 2), and visits 3 (weeks 3–4), 4 (weeks 6–8), and then every second visit (every 6 weeks) until study end. Changes in BPI individual item scores and the Pain severity and Pain Interference scale scores were compared between Z and placebo groups. Post baseline missing data were replaced by Last Observation carried forward (LOCF). Changes in scores were examined in the subgroup of patients with ECOG performance status scores of ≥ 1. Results: For patients ECOG score of ≥ 1 at baseline (n = 227), Z was superior to placebo in the BPI items ‘pain at its worst’, ‘pain at its least’, ‘pain on average’, and the Pain Severity scale at visit 4 (weeks 6–8) and visit 6 (week 12) (p < 0.05). At visit 8 (weeks 18–20) Z was superior to placebo on the ‘pain on average’ item and Pain Severity scale (p < 0.05). For visit 12 (weeks 30–32) and for LOCF Z was superior to placebo for ‘pain at its worst’, ‘pain on average’ and Pain Severity (p < 0.05 for all). For the ‘pain right now’ item there were statistically significant differences at visits 4, 6, 12 and LOCF (p < 0.05). In addition, at visit 6 (week 12), Z was superior to placebo for changes in items assessing interference with: sleep, general activities, mood, walking and enjoyment of life (p < 0.05). Z was also superior to placebo for ‘interference with general activities’ at visits 12 (weeks 30–32), 16 (weeks 42–44) and LOCF (p < 0.05). Conclusion: Results suggest Z provides significant benefit in pain reduction and Z may have an impact on patients’ daily activities which may lead to a preservation of patients’ functionality.

[Table: see text]

Higher persistency with i.v. bisphosphonates in patients with bone metastasis

18623

Background: Patients with advanced cancer are at high risk for bone metastasis, leading to accelerated bone resorption and skeletal related events and a loss of autonomy. Bisphosphonates can reduce the risk of skeletal complications, provided that patients remain on therapy. The aim of our study was to compare the persistency rates of oral and intravenous (i.v.) bisphosphonate therapy regimes in patients with bone metastasis in a natural setting. Methods: Claims data from a German sickness fund (2002 to 2004) were used. Patients had ICD-10 diagnoses for bone metastasis, were receiving oral or i.v. bisphosphonates (exclusively with a specific approval for patients with bone metastasis) and could be followed-up at least 7 months after their index prescription. Persistency was defined as the continuous prescription refill without an interruption of longer than 30 days. Medication Possession Ratio (MPR) was calculated after 180 days of follow-up and Kaplan Meier curves were analyzed. Results: We identified 233 patients who met our inclusion criteria, 85.8 % getting i.v. and 14.2 % getting oral therapy regime (prescribed bisphosphonates: zoledronate (60.5 %), pamidronate (27.5 %), clodronate (18 %), ibandronate (13.7 %)). Most of them had breast cancer (46.4 %) or prostate cancer (31.8 %). By the end of 6 months, only 64.4 % of all patients on bisphosphonate therapy were defined to be compliant, as they had a MPR ≥ 0.8 (36.4 % of the oral group and 92 % of the i.v. group). Kaplan Meier curves for oral and i.v. bisphosphonate therapy regime differed significantly (p = 0.0012). Conclusions: Persistency rates with oral bisphosphonates are significantly lower than with i.v. bisphosphonates in patients with bone metastasis (36.4 % vs. 92 %). Given the importance of preventing skeletal complications, persistency must be considered when selecting the optimum treatment regime for patients with advanced cancers.

[Table: see text]

Effect of tegaserod on work and daily activity in irritable bowel syndrome with constipation

BACKGROUND

Tegaserod is a promotility agent with proven efficacy and safety in patients with irritable bowel syndrome with constipation.

AIM

To assess tegaserod's effect on work productivity and daily activity.

METHODS

Women, 18-65 years old and meeting Rome II criteria for irritable bowel syndrome with constipation, were randomized to a double-blind, placebo-controlled, multicentre study of tegaserod 6 mg b.d. or placebo. Productivity loss and daily activity impairment because of irritable bowel syndrome were measured with the Work Productivity and Activity Impairment questionnaire for irritable bowel syndrome, modified to exclude diarrhoea as a symptom. Assessments were made at baseline, weeks 2 and 4.

RESULTS

A total of 2660 women were randomized and, of these, 1675 [tegaserod (n = 1363), placebo (n = 312)] were employed and completed Work Productivity and Activity Impairment for irritable bowel syndrome questionnaires. Compared with placebo, tegaserod significantly reduced work and daily activity impairment at weeks 2 and 4. Tegaserod reduced absenteeism by 2.6% (P = 0.004), presenteeism by 5.4% (P < 0.0001), overall work productivity loss by 6.3% (P < 0.0001), and activity impairment by 5.8% (P < 0.0001) at week 4 (vs. baseline). Assuming a 40-h workweek, tegaserod reduced work productivity loss by 2.5 h/week.

CONCLUSIONS

Tegaserod significantly reduced work productivity loss and daily activity impairment at 2 weeks, and this benefit was maintained at 4 weeks.

Irritable bowel syndrome in the United States: prevalence, symptom patterns and impact

BACKGROUND

The impact of irritable bowel syndrome, a gastrointestinal motility disorder, is underestimated and poorly quantified, as clinicians may see only a minority of sufferers.

AIM

To determine the prevalence, symptom patterns and impact of irritable bowel syndrome in the US.

METHODS

This two-phase community survey used quota sampling and random-digit telephone dialing (screening interview) to identify individuals with medically diagnosed irritable bowel syndrome or individuals not formally diagnosed, but fulfilling irritable bowel syndrome diagnostic criteria (Manning, Rome I or II). Information on irritable bowel syndrome symptoms, general health status, lifestyle and impact of symptoms on individuals' lives was collected using in-depth follow-up interviews. Data were also collected for healthy controls identified in the screening interviews.

RESULTS

The total prevalence of irritable bowel syndrome in 5009 screening interviews was 14.1% (medically diagnosed: 3.3%; undiagnosed, but meeting irritable bowel syndrome criteria: 10.8%). Abdominal pain/discomfort was the most common symptom prompting consultation. Most sufferers (74% medically diagnosed; 63% undiagnosed) reported alternating constipation and diarrhoea. Previously diagnosed gastrointestinal disorders occurred more often in sufferers than non-sufferers. Irritable bowel syndrome sufferers had more days off work (6.4 vs. 3.0) and days in bed, and reduced activities to a greater extent than non-sufferers.

CONCLUSIONS

Most (76.6%) irritable bowel syndrome sufferers in the US are undiagnosed. Irritable bowel syndrome has a substantial impact on sufferers' well-being and health, with considerable socioeconomic consequences.

Costs of care for irritable bowel syndrome patients in a health maintenance organization

OBJECTIVES

The aims of this study were: 1) to determine the total costs of care and costs related to lower GI-related problems for patients who received a diagnosis of irritable bowel syndrome (IBS), and 2) to compare them to age- and sex-matched population controls and patients treated for inflammatory bowel disease (IBD) or gastroesophageal reflux disease (GERD).

METHODS

Use and cost data were obtained through the computerized information systems of a large staff-model health maintenance organization on three groups of patients diagnosed in 1994 or 1995 with IBS, IBD, or GERD; and an age- and sex-matched control group of patients without any of these listed diagnoses. The IBS patient group was compared to the three comparison groups on components of total and IBS-related costs.

RESULTS

Total costs of care for IBS patients were 49% higher than population controls during the year starting with the visit at which IBS patients were identified. In the index year, every component of total costs except inpatient care was significantly higher for IBS patients than for population controls. The costs of care for lower GI problems were significantly higher for patients with IBS than for population controls across a range of services. However, only 33% of the difference in total costs of care between IBS patients and population controls was due to lower GI-related services in the index year. In the subsequent years, lower GI-related services accounted for 18% and 20% of the total cost difference between IBS patients and population controls. The total costs of care as well as the components of costs of care were generally higher for IBD patients than for IBS patients, but were comparable for GERD and IBS patients.

CONCLUSIONS

Patients with IBS show sustained increases in health care costs relative to population controls for both lower GI services and care unrelated to lower GI problems. However, the majority of the excess in health care costs resulted from medical care not directly related to lower GI problems.