Download Aromatase Inhibition and Breast Cancer - W. Miller, R. Santeen (Marcel Dekker, 2001) WW PDF

TitleAromatase Inhibition and Breast Cancer - W. Miller, R. Santeen (Marcel Dekker, 2001) WW
TagsMedical
LanguageEnglish
File Size2.0 MB
Total Pages328
Document Text Contents
Page 2

Aromatase Inhibition
and Breast Cancer

Page 164

Health Economics Aspects of Endocrine Therapy 147

utility values from focus groups of patients with breast cancer and derived costs
from Medicare charges and from a single center�s accounting data. The principal
results are presented in Table 1. Based on their analysis, tamoxifen and chemo-
therapy are equally cost effective in premenopausal women who are estrogen re-
ceptor (ER) positive and node negative. Combination therapy may offer an attrac-
tive option for low-risk (node-negative) women, but it is of less value than either
tamoxifen or chemotherapy alone. In node-negative, ER-negative and node-posi-
tive, ER-negative women, only chemotherapy provides a cost-effective treatment.
Although in node-positive, ER-positive patients, tamoxifen is the most cost-ef-
fective option.

This model was the first to describe the relationship between costs and ef-
fects of adjuvant therapies in premenopausal women. The cost-utility results were
highly sensitive to estimates of the relative risk of recurrence, the time frame
considered, and the relative reduction in risk expected from treatment. Interest-
ingly, changes in drug costs did not affect results significantly. The model is based
on old data and would need to be updated in view of more recent trial results.
Specifically, the model used efficacy estimates for tamoxifen that are signifi-
cantly lower than the most recent efficacy data from the EBCTCG metaanalysis
(7). Therefore, it is likely that the model underestimates the value of tamoxifen
and combination therapy treatment scenarios as compared to chemotherapy in
this population (8).

2. Chemoendocrine Therapy Versus Endocrine Therapy in
Postmenopausal Women

Goldhirsch et al. (9) conducted an analysis of quality-time without symptoms and
toxicity (Q-TWiST) to assess the balance between quality of life and survival in
postmenopausal women receiving either adjuvant tamoxifen or combination
therapy. In the Q-TWiST method, treatment duration is partitioned into phases
according to the level of toxicity and clinical events: TOX (toxicity during adju-
vant therapy), REL (relapse), and TWiST (time without symptoms and toxicity).

TABLE 1 Cost per QALY in Premenopausal Women

Tamoxifen +
Tamoxifen ($) Cyclophosphamide ($) cyclophosphamide ($)

N� ER+ 11,440 11,370 33,100
N� ER� 214,000 4,970 186,200
N+ ER+ 4,330 9,230 14,750
N+ ER� 57,800 4,890 80,700

Source: Adapted from Ref. 5.

Page 165

148 Wait

A range of theoretical utility scores, from 0 to 1, is then assigned to each phase of
treatment. The final analysis is expressed as an equation, in which:

Q-TWiST = TOX(ut) + REL(ur) + TWiST(uw)
The authors based their clinical data on results from the Ludwig Trial III, which

looked at postmenopausal node-positive women receiving either tamoxifen or
tamoxifen combined with chemotherapy for 1 year after mastectomy. Results fa-
vored the chemoendocrine arm, and the authors concluded that the increased toxic-
ity associated with combination therapy was outweighed by the survival benefit it
allowed when compared to tamoxifen alone in this patient population.

Gelber and his colleagues updated the model in 1996 with results from nine
trials involving 39,000 patients (10). After 7 years of follow-up, the toxic effects
of chemotherapy were barely balanced by the modest increase in relapsefree and
overall survival. None of the utility values tested yielded more Q-TWiST for pa-
tients undergoing chemotherapy plus hormonal therapy as compared to patients
receiving tamoxifen alone. These results are much more ìhumblingî for chemo-
therapy advocates than the previous analyses had suggested (8).

The Q-TWiST analyses do have their inherent weaknesses. The model pro-
vides an oversimplification of the phases of treatment with regard to toxicity, and
it equates ìutilityî or quality of life with toxicity, precluding any consideration
for other impacts of treatment and disease on patientsí quality of life. Also, no
costs are included in the model. Despite these limitations, the Q-TWiST method
is highly amenable to the study of adjuvant therapies, as it allows for a toxicity-
adjusted survival analysis and provides easily interpretable data for decision mak-
ing.

Another quality of life analysis in postmenopausal women receiving endo-
crine therapy was undertaken by H¸rny et al. (11). The authors conducted a pro-
spective quality-of-life assessment alongside the EBCTG Trial VII. They used a
combination of linear analogue scales for five distinct quality-of-life items. The
study found that patients receiving tamoxifen had higher scores compared to pa-
tients receiving chemoendocrine therapy. However, these differences disappeared
over time as patients adapted to the toxicity of chemotherapy over the duration of
the trial. Patient scores also reflected their anticipation of treatment. An impor-
tant limitation of this study is the choice of quality-of-life instrument. Little is
known of its psychometric properties, and extrapolation of results to different
study settings is thus difficult.

The only economic evaluation found in the literature was a cost-utility analy-
sis carried out by Hillner and Smith (12). The authors followed a similar model-
ing approach as their previous model (5) to assess the cost utility of adjuvant
tamoxifen versus tamoxifen combined with chemotherapy in postmenopausal
women. Using clinical data on the 1992 EBCTG trial results, the authors found a
cost utility of $58,000/QALY for chemoendocrine therapy; however, this ratio

Page 327

308 Index

Study 024, 126
Study 1033 IL/0029, 125
Surrogate biomarkers, 173
SYBR-Green II, 202

Tamoxifen, 106
vs. AG, 33
with AG, 33

phase III trials, 8�9
vs. anastrozole, 33
with anastrozole, 33�34
vs. aromatase inhibitors, 134�135
with aromatase inhibitors, 245�246
chemopreventive effects, 163�165
vs. chemotherapy, postmenopausal

patients, 228
with chemotherapy, 124, 137
clearance, 95
cost-effectiveness, premenopausal

patients, 147
dependence, 171
dose-response effects, tumor growth,

233
economics, 146
elderly, 103, 228

ER, 38
endometrial cancer, 228
ER, 40�41, 228, 233
ER-negative, PgR-positive tumor,

135
vs. exemestane, postmenopausal,

126
vs. fadrozole, 33
failed (see Failed tamoxifen)
vs. formestane, 33

phase III trials, 8�9
Ki-S1, 42
vs. letrozole, postmenopausal, 125�126
with letrozole, 33�34
mastectomy, 121
nonresponders, 36�37
PgR, 233
prostate cancer, 272�273
pubertal gynecomastia, 262
quality of life, 164�165
responders, 36
stroke, 228

[Tamoxifen]
vs. tamoxifen with anastrozole, post-

menopausal, 125
toremifene, 21
treatment duration, 124�125
treatment outcome, 109�111
tumor volume, 109
uterus, 233�236

Taxanes, 22
elderly, 25

Testolactone, pubertal gynecomastia,
262�263

Testosterone, 280�281
Thelarche, premature, 269
Therapeutic index, chemoprevention,

166�167
Third-generation aromatase inhibitors

vs. megestrol acetate, failed tamoxifen,
32

vs. tamoxifen, 90�91
see also Anastrozole; Exemestane;

Letrozole; Vorozole
Thymidylate synthase, 24
Thyroxine, AG, 5
Time to die (TDD), 64
Tissue processing, aromatase, immunohis-

tochemical assessment, 195
Toremifene,

prostate cancer, 273
tamoxifen, 21

Total optical density (OD), aromatase,
196

Trastuzumab, 24
Trial 0027, 58
Triazole inhibitors, 8
Tumor necrosis factor-α (TNF-α), 281
Tumors,

decellulariation, 249�250
recurrent

ER status, 38�39

Uterus,
letrozole, 234�236
tamoxifen, 233�236

Velcro chest binders, 261�262
Vinca alkaloids, elderly, 25

Page 328

Index 309

Visceral disease, progestins, vs. aromatase
inhibitors, 36

Visceral metastases,
age factors, 85�86
letrozole, 59

Vorozole, 8
vs. aminoglutethimide, 56
chemoprevention, clinical trials, 168�

169

clinical overview, 52
mammary tumor incidence, 168
vs. megestrol acetate, 55
preclinical models, 167�168
vs. tamoxifen, 42

YM511, 8

Zoladex, 132

Similer Documents