Kim Bergstrom, PharmD

To evaluate chemotherapy regimen utilization in patients with non—small cell lung cancer (NSCLC) treated in US community oncology practices, to examine the relationship between evidence-based guideline adherence and the follow-up monitoring period (FUMP) over 1.5 years, and to understand the relative costs of commonly administered chemotherapy regimens.

Using a large US medical oncology clinical database derived from a proprietary web-based drug dispensing technology, we identified adult patients with NSCLC who started adjuvant therapy for early-stage disease or first-line therapy for advanced and metastatic disease from July 1, 2009, through June 30, 2010. Adjuvant or first-line regimen utilization and the FUMP within 1.5 years were analyzed with respect to national evidence-based guideline adherence. Costs for commonly administered regimens based on 2010 Medicare reimbursement were compared.

A total of 3505 patient treatment regimens were included in this study. Rates of guideline adherence were 75.0% and 61.3% for the first-line and the adjuvant treatment groups, respectively (

 <.0001). Treatment with guidelinebased regimens correlated with a significantly longer FUMP in the first-line treatment groupcompared with non—guideline-based regimens (

 = .005). Regimen costs for the top 11 regimens in the adjuvant and first-line treatment settings varied greatly. Low-cost regimens were prescribed more commonly.

Rates of guideline adherence were significantly higher in the first-line than in the adjuvant NSCLC treatment group. First-line treatment with guideline-based regimens correlated with an extended FUMP for advanced NSCLC patients.

Retrospective analysis of a large US medical oncology clinical database indicated that rates of guideline adherence for community oncology practices in adjuvant and first-line treatment of non—small cell lung cancer (NSCLC) were 61.3% and 75.0% (

		First-line treatment with guideline-adherent regimens correlated with a longer followup monitoring period in advanced NSCLC.

		Regimen standardization would increase rates of guideline adherence and lower treatment costs.

		Onmark Regimen Profiler is a financial tool to help clinicians to compare costs of treatments at the regimen level.

With the exception of skin cancer, lung cancer is the most common malignancy and is the leading cause of cancerrelated mortality worldwide.

 An estimated 222,520 Americans were diagnosed with lung cancer in 2010, while an estimated 157,300 died of the disease.

 There are 2 major subtypes of lung cancer: non—small cell lung cancer (NSCLC) and small cell lung cancer (SCLC). The majority (85%-90%) of lung cancers fall into the NSCLC subtype; a smaller percentage (10%-15%) are classified as SCLC or other cell types (up to 5%).2 For NSCLC, the 5-year survival rate is 46.9%, 26.1%, 8.4%, and 1.6% for stages I, II, III, and IV, respectively.

National treatment guidelines (NTGs) provide evidence-based recommendations for treatment, and are available from a number of organizations, including the National Comprehensive Cancer Network,

 the American Society of Clinical Oncology,

Treatment for NSCLC has evolved over the past 20 years to include chemotherapy, surgery, and radiation therapy. Chemotherapy plays an important role in both early and advanced disease. The treatment for early-stage NSCLC has evolved to include multimodality therapy utilizing chemotherapy and/or radiation therapy in addition to surgery, if the tumor is operable, in an effort to improve survival. Advanced NSCLC is often managed with chemotherapy or chemoradiation, which has been shown to prolong survival.

Doublet chemotherapy regimens, particularly platinum-containing regimens, have emerged as standard treatment and are recommended by NTGs for neoadjuvant, adjuvant, initial, or first-line metastatic treatment for NSCLC.

 Cisplatin-containing regimens are morecommonly recommended in NTGs in the adjuvant setting than carboplatin-containing or other chemotherapy combinations.

 A meta-analysis demonstrated that, compared with surgery alone, cisplatin-based regimens are superior to non—cisplatin-containing regimens, based on a 13% reduction in the risk of death and a 5% absolute improvement in 5-year survival for patients receiving cisplatin-based regimens and a 15% increase in the risk of death for patients receiving non–cisplatin-containing regimens.

 However, no specific platinumbased cytotoxic combination is clearly superior.

Use of chemotherapy varies greatly across the United States.

 Many different regimens are available, with varying efficacy, side effect profiles, and costs.

There is increasing interest in better understanding the value obtained from specific cancer treatments. Value can be determined by examining quality (using physician adherence to evidence-based NTGs as a surrogate), treatment outcome (eg, survival), and cost of different cancer treatment options.

 Previous studies have demonstrated significant variability in adherence to guideline-driven care in oncology.

 When attempting to define the benefits of clinical practice guidelines, evidence for improvement in 1 of the 3 major outcomes is used as a primary determinant for the recommendations: (1) survival, (2) qualify of life, and 3) cost-effectiveness.

The purpose of this retrospective study was to compare adjuvant and first-line NSCLC treatments prescribed by US community oncologists with the NTGs for this condition. Treatment regimen cost and the follow-up monitoring period (FUMP) as a surrogate for survival were examined too. Quality-of-life measurements, which have been shown to benefit little from chemotherapy in this population, were beyond the scope of this study.

Data were collected from Lynx Mobile, McKesson Specialty Health’s web-based drug inventory management technology. The Lynx Mobile data warehouse contains data from oncology patients treated in more than 750 US community oncology practice sites since 2001. The data warehouse contains patient clinical and demographic data as well as drug transactional data.

Eligible study patients were 18 to 80 years old with a diagnosis of NSCLC (utilizing 

International Classification of Diseases, Ninth Revision, Clinical Modification

 billing codes of 162.0-162.9 and NSCLC as a clinical diagnosis supplied by the clinician in the Lynx Mobile system) and with treatment for NSCLC started between July 1, 2009, and June 30, 2010. Patients were included if they started neoadjuvant/adjuvant treatment (defined as the adjuvant treatment group) for stage I-IIIa disease, or if they started first-line treatment including initial treatment of locally advanced disease and first-line metastatic treatment (defined as the first-line treatment group) for stage IIIb-IV disease. Patient data, including diagnosis, tumor stage, height, weight, and drug-related transactional data (doses and days of administration), were collected. Additional inferred information was added to the data set by a team of clinicians at McKesson, including the patients’ treatment regimen (utilizing a regimen library matched to drug dose and dosing patterns) and treatment line of therapy (LOT) (eg, neoadjuvant, adjuvant, initial/induction, firstline metastatic) using LOT rules and staging information.

Treatment was analyzed based on treatment intent. Patients were excluded from the study if their sex was unknown or if they received only 1 cycle of chemotherapy. Patients were excluded from the FUMP analysis if the FUMP was shorter than 21 days. If patients were given more than 1 regimen for the same treatment (eg, 2 regimens given in the adjuvant setting), only the first regimen was eligible for data analysis.

Treatment regimen cost information was derived from Onmark Regimen Profiler, a web-based tool developed by McKesson Specialty Health that evaluates, on a pro forma basis, regimen-associated costs for different payers including Medicare. Regimen costs were based on the first quarter 2010 Medicare Drug Fee Schedule and the 2010 National Average Procedure Fee Schedule. The regimen cost was the Medicare-allowable cost for a regimen in this study. Only adjuvant and first-line regimens were analyzed. First-line regimen costs assumed 4 cycles of therapy. Typical Evaluation & Management, administration procedures (such as injection method and infusion time), and basic lab tests were included. Dose calculations were based on an average patient body surface  area of 1.7 m

 and weight of 70 kg. Pegfilgrastim (6-mg injection) and an antiemetic regimen (palonosetron 0.25 mg [intravenous infusion] and dexamethasone 20 mg) were included in the costs for all regimens associated with a high risk of febrile neutropenia or a high to moderate risk of emesis, respectively.

Rules to Define Regimen Category, Line of Therapy, Guideline Adherence, and Follow-up Monitoring Period

Regimens containing either carboplatin or cisplatin were categorized as carboplatin- or cisplatincontaining regimens, respectively.

Adjuvant patients were new to the practice, had stage I-IIIA disease, received therapy consistent with neoadjuvant or adjuvant treatment, and did not receive chemotherapy again within 90 days. First-line patients were either new to the practice, had stage IIIB or IV disease, had no previous chemotherapy, and received a regimen consistent with initial or first-line treatment, 

were patients who had been previously treated with an adjuvant regimen and then received new therapy with a gap of 

Current staging, when available, was one of the factors used to determine appropriate LOT.

Maintenance treatment and second-line or higher metastatic treatment were not included in this study.

Guideline adherence refers to regimens referenced in evidence-based NTG sources during the time of the study.

The FUMP was examined for the first-line patients only and was measured from the first date of first-line treatment to last clinical contact recorded in our database by the end of 1.5 years of monitoring. These data were accessed on January 13, 2012.

The rates of guideline adherence for adjuvant and firstline treatments, the percentages of carboplatin- and cisplatin-containing regimens, and the regimen drug counts were evaluated statistically by x

 The FUMP was evaluated by 1-way analysis of variance (ANOVA) test with MedCalc software.

 value of <.05 was considered significant.

Between July 1, 2009, and June 30, 2010, a total of 3505 NSCLC patient regimens met the inclusion criteria. Median patient age was 68 years (SD 8.7 years). These patients were treated by 336 community oncology practices across 41 states in the United States.

A total of 661 patients received adjuvant treatment. Of this group, 24.5% (n = 162) had valid staging information in the database. The proportions of patients with stage IIIA, II, and I disease were 46.9%, 31.5%, and 21.6%, respectively. More than 80% (n = 2844) of the patients in the sample received first-line treatment for advanced/metastatic disease. Among the patients in the first-line treatment group,37.1% (n = 1056) had valid staging information. Two cases in the adjuvant group and 3 in the first-line group had invalid staging information and were eliminated from valid staging analysis. 

 shows the demographic characteristics of the patients.

In this study, about 70% of patients had their body surface area, body weight, and height data recorded. When these data were absent, assumptions about drug dose and regimen selection were based on default values.

Articles

US community oncologists treating NSCLC were significantly more likely to be guideline adherent when providing first-line rather than adjuvant treatment.

Kim Bergstrom, PharmD

Articles

Rates of Guideline Adherence Among US Community Oncologists Treating NSCLC