Economic Analysis of Targeting Chemotherapy Using a 21-Gene RT-PCR Assay in Lymph-Node"Negative, Estrogen-Receptor"Positive, Early-Stage Breast Cancer

Published Online: May 01, 2005
John Hornberger, MD; Leon E. Cosler, PhD, RPh; and Gary H. Lyman, MD, MPH, FRCP (Edin)

Objective: To appraise the economics of a recurrence score (RS), based on an assay that predicts distant recurrence-free survival in lymph-node-negative (LN-), estrogen-receptor-positive (ER+) patients with early-stage breast cancer receiving tamoxifen.

Study Design: Cost-utility analyses using a decision analytic model.

Methods: Using a Markov model, we forecast overall survival, costs, and cost effectiveness of using the RS in patients classified as having low or high risk of distant recurrence based on National Comprehensive Cancer Network (NCCN) clinical guidelines. Data from a large multicenter clinical trial (NSABP B-14) were analyzed to derive risk classification based on guideline criteria and RS assignments. Efficacy of adjuvant chemotherapy (CT) on distant recurrence-free survival (DRFS) was based on published metaanalyses of CT trials. The analysis took a societal perspective, considering survival, quality of life, and relevant costs.

Results: Fifty-three patients (8%) were classified as having low risk of distant recurrence by NCCN guidelines and the RS reclassified 15 of these patients (28%) to an intermediate/high-risk group. The remaining 615 patients (92%) were classified at high risk of distant recurrence by NCCN guidelines and the RS reclassified 300 of these patients (49%) to a low-risk group. Among a hypothetical cohort of 100 patients, RS is predicted on average to increase quality- adjusted survival by 8.6 years and reduce overall costs by $202 828. RS was cost saving in more than two-thirds of probabilistic simulations, with cost effectiveness most influenced by the propensity to administer CT based on RS results, and by the proportion of patients at low risk as defined by NCCN guidelines.

Conclusions: The RS predicts more accurately than current guidelines recurrence risk in LN-, ER+ patients with early-stage breast cancer. If applied appropriately, the assay is predicted to increase quality-adjusted survival and save costs.

(Am J Manag Care. 2005;11:313-324)

More than 210 000 women in the United States are diagnosed each year with breast cancer.1 Breast cancer remains among the most common cancers in women, and the most common cause of death among women between the ages of 40 and 79.1

Several large randomized clinical trials demonstrated the benefit of hormonal therapy in patients with estrogen- receptor-positive (ER+) early-stage breast cancer (ESBC).2 An important decision for a patient with lymph-node-negative (LN-), ER+ ESBC is whether to also undergo adjuvant chemotherapy after primary surgery to prevent or delay distant recurrence.3-7 Consensus guidelines endorse the addition of adjuvant chemotherapy for LN-, ER+ cancer for patients up to 70 years old, or older if they are medically fit.8-11 Experts also recommend against routine use of adjuvant chemotherapy for small tumors (<1 cm) or for small tubular or mucinous tumors.12

Enhanced public health efforts to detect breast cancer, such as mammographic screening, have increased early-stage detection.13,14 The success of this campaign has naturally resulted in physicians and patients increasingly facing a complex question: do the benefits of adjuvant chemotherapy outweigh the medical risks and known adverse effects on quality of life?15 That this question is difficult to answer is supported by recent evidence showing wide variation in the propensity to prescribe adjuvant chemotherapy, a variation that cannot be explained by characteristic risk factors such as age, tumor size, and histology.10,16-21 An active area of oncology research therefore is identifying additional reliable predictors of recurrence in ESBC that would assign risk more accurately and help guide the decision to prescribe adjuvant chemotherapy.22-27

A 21-gene reverse transcriptase-polymerase chain reaction (RT-PCR) assay (Oncotype DX Breast Cancer Assay, Genomic Health, Inc, Redwood City, Calif) generates an individualized "recurrence score" that is derived from a proprietary algorithm. The recurrence score has been prospectively validated as a predictor of 10-year distant recurrence-free survival (DRFS) in patients with LN-, ER+ ESBC.28 The investigators obtained tissue samples from 668 LN-, ER+ patients who were treated with tamoxifen in the National Surgical Adjuvant Breast Cancer Project (NSABP) B-14 clinical trial from 1982 through 1988 and whose outcomes have been tracked over time by NSABP sites. The recurrence score accurately classified patients into low and high risk of DRFS (P < .001).28 When the recurrence score was examined together with age and tumor size in a multivariate analysis, only recurrence score remained a significant predictor of DRFS at 10 years (P < .001).28 The study also confirmed the relatively poor reliability of current risk stratification. Twelve of 53 (22%) patients initially classified by National Comprehensive Cancer Network (NCCN) criteria as low risk were reclassified by the recurrence score as intermediate risk and had a 10-year DRFS equal to 82% (95% confidence interval [CI] 60%-100%); 3 (6%) patients were reclassified as high risk and had a 10-year DRFS equal to 57% (95% CI 1%-100%) (Figure 1).29 Conversely, 300 of 615 (49%) patients assigned by NCCN criteria as high risk for recurrence were reclassified by the recurrence score as low risk and had a 10-year DRFS equal to 92% (95% CI 89%-95%).


Although the recurrence score has been subjected to clinical validation, questions necessarily arise about its affordability and factors that would influence its appropriate use. The purposes of this study were (1) to estimate the incremental benefits, costs, and cost effectiveness of using the recurrence score to better assign risk of DRFS associated with ESBC, and (2) to assess the factors that most influence potential benefits and efficient use of the recurrence score.


This evaluation focused on the recurrence score and not other RT-PCR or genomic assays under development, because the analyses depend critically on the quality of evidence and detailed findings of the validation studies described herein. We present DRFS, overall survival (OS), and relevant costs associated with use of the recurrence score to reclassify risk of recurrence compared with risk classification using current NCCN guidelines alone. Outcomes were evaluated using a common framework for health economic appraisals, called the Markov model (Figure 2), which provides a convenient way of modeling prognosis.30 In a Markov model, a patient may be in 1 of a finite number of states of health, and events of interest are modeled as transitions from one state to another. For each state, analysts assign a utility as an adjustment factor for quality of life. Utility weights typically range from 0 to 1, in which 0 represents a state as bad as death, 1 represents perfect health, and the values between 0 and 1 represent degrees between these extremes. The contribution to total utility, commonly referred to as quality-adjusted life years (QALYs), of a particular state consists of the length of time spent in a state multiplied by the utility of that state.


Two scenarios were considered involving representative patients with LN-, ER+ ESBC who were expected to receive 5 years of hormonal therapy. In scenario 1, patients were classified by NCCN as low risk (eg, tumor size <1 cm), would therefore not receive chemotherapy, and would have a predicted 10-year probability of recurrence of 7.8% based on analyses of NSABP B-14 data (Table 1). In scenario 2, patients were classified by NCCN as high risk (eg, tumor size >2 cm), would therefore be recommended to receive chemotherapy, and would have a predicted 10-year probability of recurrence of 21.9%.


The recurrence score was assumed to reclassify recurrence risk independent of NCCN risk criteria, with the probability of reclassification based on results of the NSABP B-14 data. In the base case, we assumed that all patients assigned as intermediate/high risk by the recurrence score would undergo chemotherapy and all patients assigned as low risk by the recurrence score would not receive chemotherapy. We explored in sensitivity analysis the implications of different probabilities of chemotherapy use based on risk classification by the recurrence score. The model tracked rates of recurrence and death, as shown in the outcome subtree of the decision tree in Figure 2.

Probability of Risk Reclassification

The NSABP provided us with analyses of B-14 data to determine the risk of recurrence based on NCCN criteria and on the recurrence score results (Figure 1 and Table 2).29 For the 53 (7.9%) women assigned as having low risk for recurrence based on NCCN criteria, NSAPB B-14 showed a 28% (95% CI 16.8%-42.4%) probability of reclassification to intermediate/high risk (15 of 53 women) with the recurrence score. For the other 615 (92.1%) women assigned as having high risk for recurrence based on NCCN criteria, NSABP B-14 showed a 49% (95% CI 44.5%-52.6%) probability of reclassification to low risk (300 of 615 women) based on the recurrence score.


Risk of Recurrence and Death

Annual risks of recurrence and survival were obtained from published meta-analyses of clinical trials (Table 1).31 The relative risk reduction of distant recurrence associated with chemotherapy plus tamoxifen versus tamoxifen only is approximately 30%, and is assumed to be the same regardless of NCCN risk classification.31 Studies are ongoing to assess the correlation of tumor gene expression and response to chemotherapy.35 Gianni et al recently reported that gene expression profiles of paraffin-embedded core biopsy tissue predicted response to chemotherapy in patients with locally advanced breast cancer,36 which was subsequently confirmed in an NSABP study.37 In the base case model, we assumed a 15% risk reduction associated with chemotherapy if the recurrence score falls in the low-risk category and a 45% risk reduction associated with chemotherapy if the recurrence score falls in the intermediate/ high-risk category.

After recurrence, the chemotherapeutic regimen, probability of response, and risk of further disease progression depend on Her2/neu status, assumed to be positive in 25% of cancers.32 Regardless of response, the rate of disease progression is lower for patients with Her2/neu-positive tumors receiving combination trastuzumab/paclitaxel than for patients with Her2/neunegative tumors receiving paclitaxel monotherapy. The annual probability of death after progression is approximately 40%.32

Quality of Life

PDF is available on the last page.


Recommended Reading

No Result Found