Adjunctive traditional Chinese medicine therapy improves survival in patients with advanced breast cancer: A population-based study
This study was based in part on data from the National Health Insurance Research Database provided by the Bureau of National Health Insurance, Department of Health and managed by National Health Research Institutes. The interpretations and conclusions contained herein do not represent the opinions of the Bureau of National Health Insurance, Department of Health, or the National Health Research Institutes.
Abstract
BACKGROUND
Traditional Chinese medicine (TCM) is one of the most common complementary and alternative medicines used in the treatment of patients with breast cancer. However, the clinical effect of TCM on survival, which is a major concern in these individuals, lacks evidence from large-scale clinical studies.
METHODS
The authors used the Taiwan National Health Insurance Research Database to conduct a retrospective population-based cohort study of patients with advanced breast cancer between 2001 and 2010. The patients were separated into TCM users and nonusers, and Cox regression models were applied to determine the association between the use of TCM and patient survival.
RESULTS
A total of 729 patients with advanced breast cancer receiving taxanes were included in the current study. Of this cohort, the mean age was 52.0 years; 115 patients were TCM users (15.8%) and 614 patients were TCM nonusers. The mean follow-up was 2.8 years, with 277 deaths reported to occur during the 10-year period. Multivariate analysis demonstrated that, compared with nonusers, the use of TCM was associated with a significantly decreased risk of all-cause mortality (adjusted hazards ratio [HR], 0.55 [95% confidence interval, 0.33-0.90] for TCM use of 30-180 days; adjusted HR, 0.46 [95% confidence interval, 0.27-0.78] for TCM use of > 180 days). Among the frequently used TCMs, those found to be most effective (lowest HRs) in reducing mortality were Bai Hua She She Cao, Ban Zhi Lian, and Huang Qi.
CONCLUSIONS
The results of the current observational study suggest that adjunctive TCM therapy may lower the risk of death in patients with advanced breast cancer. Future randomized controlled trials are required to validate these findings. Cancer 2014;120:1338–1344. © 2014 American Cancer Society.
INTRODUCTION
Breast cancer is the most prevalent malignant tumor in female patients worldwide and the leading cause of death in women with cancer, accounting for approximately 1.38 million new diagnoses and 458,400 deaths every year.1 Although patient incidence and mortality are decreasing in some countries, the mortality rate of advanced breast cancer remains high.2 The current treatment for patients with advanced breast cancer includes surgery, radiotherapy, chemotherapy, and hormonal and targeted biological therapies.3 Recent studies have documented that at least 46% of patients with breast cancer receive complementary and alternative medicine treatments4, 5 to boost immune system activity, decrease disease symptoms, minimize side effects resulting from conventional treatments, increase quality of life, and even as a treatment of cancer per se.6
Traditional Chinese medicine (TCM) is one of the most widely used complementary and alternative medicine therapies used by patients with breast cancer worldwide.7, 8 Although studies have indicated that TCM facilitates the treatment of breast cancer, to the best of our knowledge, the majority of the reports on the alleviation of side effects by chemotherapy and hormonal therapy originate from laboratory findings.9-11 Patients are mostly concerned about disease mortality arising from the progression of cancer; however, to our knowledge large-scale clinical analysis of TCM on disease mortality in patients with breast cancer remains absent. In Taiwan, TCM is covered by National Health Insurance (NHI) and is a widely accepted form of medical treatment. According to the study by Lai et al, in addition to receiving conventional therapy, approximately 81.5% of patients with breast cancer in Taiwan had ever received any treatment with TCM during the 10-year study period.12 Using the population-based National Health Insurance Research Database (NHIRD), we investigated whether the combination of TCM and conventional cancer treatment affected the survival of patients with advanced breast cancer.
MATERIALS AND METHODS
Research Database
The NHI system in Taiwan is a compulsory, single-payer program that provides health coverage to nearly every individual in Taiwan. Currently, 99.6% of residents are covered by NHI.13 The Bureau of National Health Insurance allowed a random selection of 1 million representative patient records from the NHIRD (23 million insured individuals) to be distributed for research. The age, sex distribution, and premiums paid were not found to be statistically significantly different between the 1 million insured individuals and the general population.14 In the current retrospective population-based cohort study, we analyzed 1 million patients randomly selected from 23 million beneficiaries in the NHIRD between January 1, 1999 and December 31, 2010. The NHIRD contains comprehensive outpatient and inpatient information including age, sex, date of visit, International Classification of Diseases, Ninth Revision, Clinical Modification (ICD-9-CM) codes, and complete prescription records. The detailed diagnoses and treatments provided by Chinese medicine physicians were also included. This study was approved by the Joint Institutional Review Board of Taipei Medical University.
Study Population
Insured individuals of the NHI diagnosed with severe and chronic diseases that require extended treatment, such as breast cancer, may apply for a catastrophic illness certificate to be exempted from copayments. To receive a catastrophic illness certificate, the diagnosis of a patient with cancer must be confirmed by specialists and pathological reports as was the case in the current study. Once patients fulfill the above-mentioned conditions, physicians and medical institutions must provide assistance to the patient to apply for the catastrophic illness certificate. The status of catastrophic illness is then registered into the NHI card. The current study comprised women aged > 18 years who were diagnosed with primary breast cancer (ICD-9-CM code 174) between 2001 and 2010. Although the stage of disease of patients with breast cancer cannot be obtained from the ICD-9-CM codes, we identified those patients with advanced breast cancer through taxane prescriptions. We used this approach because the Bureau of National Health Insurance reimburses taxane use only for those patients with locally advanced or metastatic breast cancer. Patients with advanced breast cancer who received taxane (docetaxel or paclitaxel) treatments between January 1, 2001 and December 31, 2010 were included in the current study. The initiation of taxane treatment was defined as the index date. We obtained detailed hospitalization, outpatient, and prescription information from 2 years before the index date to the end of the follow-up period for analysis. Patients who received taxane treatment before January 1, 2001 and those with a follow-up period of < 3 months were excluded.
The baseline demographic variables of the patients were obtained from the registry for beneficiary files, and patient age was determined according to the index date. The baseline comorbidities were determined using the modified version of the Elixhauser comorbidity index15 according to the patient medical records 2 years before the index date. These comorbidities include hypertension, diabetes mellitus, congestive heart failure, stroke, chronic pulmonary disease, and liver disease. Distant metastases were defined as metastases to the lungs, liver, brain, bones, and other organs (ICD-9-CM codes 197.x, 198.0, 198.1, and 198.3-198.7). The type of treatment was considered as the treatment that a patient with breast cancer received 2 years before the index date until the end of the follow-up period. A patient undergoing both mastectomy and breast-conserving surgery was classified under mastectomy. Combination chemotherapy was defined as when patients received other chemotherapy medications in addition to taxanes.
Exposure to TCM
The NHIRD records detailed prescription information for both TCM and Western medicine including drug name, dose, start date, frequency, duration, and method of administration. All TCMs covered by the NHI were prescribed by board-certified Chinese medicine physicians who received rigorous training in both Chinese and Western medicine in medical schools and through residency programs in hospitals. These physicians must also pass the national licensing examinations. TCMs are prescribed by Chinese medicine physicians according to TCM symptom patterns.10 Patients using TCM for ≥ 30 days due to a diagnosis of breast cancer were defined as TCM users, whereas those treated for < 30 days were considered as TCM nonusers. Moreover, to observe a dose-response relationship, we further grouped TCM users into 2 groups: 1 group used TCM for 30 to 180 days whereas the other group used TCM for > 180 days.
Study Outcome
The study outcome was all-cause mortality during the 10-year follow-up. The patient's date of death was determined according to the Registry of Catastrophic Illness Database of the NHIRD. The analysis period began from the index date and ended at the death of patient, withdrawal from NHI, or the end of 2010, whichever occurred first.
Statistical Analysis
The continuous variables were summarized using means and standard deviations, whereas the categorical variables were summarized using counts and percentages. The continuous variables of the baseline characteristics of TCM users and nonusers were analyzed using a Student t test, and the categorical variables were analyzed using a chi-square or Fisher exact test. The cumulative probability of survival for TCM users and nonusers was estimated using a Kaplan-Meier estimator16 with a log-rank test17 used to compare the survival curves between the groups. The association between the baseline characteristics and mortality was estimated using a univariate Cox regression analysis18 and reported according to hazards ratios (HRs) with 95% confidence intervals (95% CIs). The variables with P values < .2 in the univariate analysis were included in the multivariable Cox proportional hazards model to identify the independent predictors of mortality. Log(-log[survival]) versus log of survival time plot was inspected to verify the proportional hazards assumption. The study endpoint was all-cause mortality. Patients who were alive at the time of last follow-up were censored at the date of withdrawal from NHI or the end of study period, whichever came first. To verify the dose-response relationship between TCM use and mortality, we treated TCM use category as a continuous variable to calculate the P value of the linear trend. Because stage of cancer is a major predictor of mortality, we conducted a sensitivity analysis limiting to patients with distant metastases to evaluate the robustness of our findings. Statistical significance was determined as P < .05. All analyses were performed using SAS statistical software (version 9.3; SAS Institute Inc, Cary, NC).
RESULTS
Of the 1 million people randomly selected from the NHIRD between 2001 and 2010, 786 female patients were aged > 18 years and received taxane treatment because of primary breast cancer. We further excluded 5 patients who received taxane treatment before January 1, 2001 and 52 patients with a follow-up period of < 3 months. Of the remaining 729 eligible patients, 614 were TCM nonusers, whereas the remaining 115 patients were TCM users, including 58 users who took TCM for 30 to 180 days and 57 users who used it for > 180 days. Characteristics such as age, urbanization, baseline comorbidity, and treatment did not appear to differ significantly between the groups (Table 1).
| TCM Users (N = 115) |
TCM Nonusers (N = 614) |
||||
|---|---|---|---|---|---|
| Characteristics | Mean (SD) or No. (%) | Mean (SD) or No. (%) | P | ||
| Age, y | 50.5 | (8.9) | 52.3 | (10.0) | .08 |
| Urbanization | |||||
| Low | 8 | (7.0) | 37 | (6.0) | .63 |
| Moderate | 24 | (20.9) | 153 | (24.9) | |
| High | 83 | (72.2) | 424 | (69.1) | |
| Comorbidity | |||||
| Hypertension | 10 | (8.7) | 83 | (13.5) | .16 |
| Diabetes mellitus | 1 | (0.9) | 25 | (4.1) | .10 |
| Congestive heart failure | 3 | (2.6) | 15 | (2.4) | >.99 |
| Stroke | 1 | (0.9) | 24 | (3.9) | .16 |
| Chronic pulmonary disease | 15 | (13.0) | 92 | (15.0) | .59 |
| Liver disease | 16 | (13.9) | 100 | (16.3) | .52 |
| Distant metastases | 41 | (35.7) | 228 | (37.1) | .76 |
| Type of treatment | |||||
| Surgery | |||||
| None | 28 | (24.4) | 191 | (31.1) | .29 |
| Breast-conserving surgery | 19 | (16.5) | 105 | (17.1) | |
| Mastectomy | 68 | (59.1) | 318 | (51.8) | |
| Combination of chemotherapy | 110 | (95.7) | 586 | (95.4) | .92 |
| Hormonal or targeted biologic therapy | 81 | (70.4) | 409 | (66.6) | .42 |
| Radiotherapy | 66 | (57.4) | 367 | (59.8) | .63 |
- Abbreviations: SD, standard deviation;TCM, traditional Chinese medicine.
The mean follow-up period for the patients was 2.8 years ± 1.9 years; 277 deaths occurred and the overall mortality rate was 38.0% during the analysis period. TCM users had 32 deaths (27.8%) and TCM nonusers had 245 deaths (39.9%). The Kaplan-Meier survival curve and log-rank test revealed a statistically significant difference between the survival curves of the groups (P < .001) (Fig. 1).
The unadjusted Cox regression analysis demonstrated a strong association between the use of TCM and a decrease in mortality (Table 2). Compared with TCM nonusers, TCM users had reduced mortality by 47% (unadjusted HR, 0.53; 95% CI, 0.37-0.77 [P < .001]). On the multivariate Cox model controlling for 6 potential confounders (age, history of congestive heart failure, distant metastases, surgery, hormonal or targeted biologic therapy, and radiotherapy), the use of TCM remained highly associated with decreased mortality (the adjusted HR of TCM users was 0.50 [95% CI, 0.35-0.73]; P < .001) (Table 2).
| Univariate Analysis | Multivariate Analysis | |||||
|---|---|---|---|---|---|---|
| Variable | HR | 95% CI | P | aHRa | 95% CI | P |
| TCM use | ||||||
| Nonusers (<30 d) | 1 | 1 | ||||
| TCM users (≥30 d) | 0.53 | 0.37-0.77 | <.001 | 0.50 | 0.35-0.73 | <.001 |
| 30-180 d | 0.58 | 0.36-0.95 | .03 | 0.55 | 0.33-0.90 | .02 |
| >180 d | 0.48 | 0.29-0.81 | .006 | 0.46 | 0.27-0.78 | .004 |
| Age (per 1 y) | 1.01 | 1.00-1.02 | .12 | 1.00 | 0.99-1.02 | .58 |
| Congestive heart failure | 1.91 | 0.98-3.72 | .06 | 1.31 | 0.67-2.56 | .44 |
| Distant metastases | 3.10 | 2.44-3.94 | <.001 | 2.44 | 1.84-3.24 | <.001 |
| Surgery | ||||||
| None | 1 | 1 | ||||
| Breast-conserving surgery | 0.42 | 0.30-0.61 | <.001 | 0.56 | 0.38-0.81 | .002 |
| Mastectomy | 0.33 | 0.25-0.43 | <.001 | 0.54 | 0.40-0.73 | <.001 |
| Hormonal or targeted biologic therapy | 1.46 | 1.10-1.92 | .01 | 1.06 | 0.80-1.41 | .69 |
| Radiotherapy | 1.36 | 1.06-1.75 | .02 | 1.56 | 1.21-2.01 | <.001 |
- Abbreviations: 95% CI, 95% confidence interval; aHR, adjusted hazards ratio; HR, hazards ratio; TCM, traditional Chinese medicine.
- a Adjusting for all variables listed.
Further analysis demonstrated a dose-response relationship between TCM use and mortality. The adjusted HRs were 0.55 (95% CI, 0.33-0.90) and 0.46 (95% CI, 0.27-0.78) for patients with TCM use of 30 to 180 days and > 180 days, respectively. The longer the duration of TCM use, the lower the mortality rate (P for trend < .001). The other independent predictors of increased mortality included distant metastases, inoperable surgical status, and radiotherapy.
Similar results were obtained when analyses were restricted to patients with distant metastases. Of the 269 patients with distant metastases, after adjusting for age, history of congestive heart failure, surgery, hormonal or targeted biologic therapy, and radiotherapy, TCM use was still associated with reduced mortality (adjusted HR 0.39; 95% CI, 0.23-0.67 [P < .001]) when compared with nonuse of TCM.
The most common prescriptions of the 115 TCM users were Jia Wei Xiao Yao San, Pu Gong Ying, and Bai Hua She She Cao, which were used by 57 TCM users (49.6%), 51 TCM users (44.3%), and 50 TCM users (43.5%), respectively (Table 3). Of the 10 most common TCMs, 3 were herbal formulae and 7 were single herbs. Cox proportional hazards regression analysis demonstrated that the 3 most effective (lowest HRs) TCM agents in reducing mortality were Bai Hua She She Cao, Ban Zhi Lian, and Huang Qi (Table 4). By contrast, the results of the current analysis revealed Dan Shen, Xiang Sha Liu Jun Zi Tang, and Ban Xia were not significantly associated with reduced mortality.
| TCM Name | Ingredients or Generic Name | Functional Classification | No. of Users | % |
|---|---|---|---|---|
| Jia Wei Xiao Yao San | Angelica sinensis, Atractylodes macrocephala, Bupleurum chinensis, Gardenia jasminoides, Glycyrrhiz auralensis, Mentha haplocalyx, Paeonia lactiflora, Paeonia suffruticosa, Poria cocos, Zingiber officinale | Harmonizing and releasing formulae | 57 | 49.6 |
| Pu Gong Ying | Taraxacum mongolicum | Heat-clearing and detoxicating | 51 | 44.3 |
| Bai Hua She She Cao | Oldenlandia diffusa | Heat-clearing and detoxicating | 50 | 43.5 |
| Huang Qi | Astragalus membranaceus | Qi-tonifying (ie, restore energy flow) | 49 | 42.6 |
| Dan Shen | Salvia miltiorrhiza | Blood-activating and stasis-resolving | 46 | 40.0 |
| Xiang Sha Liu Jun Zi Tang | Amomum villosum, Atractylodes macrocephala, Aucklandia lappa, Citrus reticulata, Glycyrrhiza uralensis, Panax ginseng, Pinellia ternata, Poria cocos, Zingiber officinale | Tonifying and replenishing formulae | 44 | 38.3 |
| Ji Xue Teng | Spatholobus suberectus | Blood-activating and stasis-resolving | 42 | 36.5 |
| Ban Zhi Lian | Scutellaria barbata | Heat-clearing and detoxicating | 40 | 34.8 |
| Gui Pi Tang | Angelica sinensis, Astragalus membranaceus, Atractylodes macrocephala, Aucklandia lappa, Dimocarpus longan, Glycyrrhiza uralensis, Panax ginseng, Polygala tenuifolia, Poria cocos, Zingiber officinale, Ziziphus jujuba, Ziziphus spinosa | Tonifying and replenishing formulae | 38 | 33.0 |
| Ban Xia | Pinellia ternata | Phlegm-resolving | 37 | 32.2 |
- Abbreviations: TCM, traditional Chinese medicine.
| Univariate Analysis | Multivariate Analysis | |||||
|---|---|---|---|---|---|---|
| Individual TCM Use (≥30 d)a | HR | 95% CI | P | aHRb | 95% CI | P |
| Jia Wei Xiao Yao San | 0.32 | 0.13–0.77 | .01 | 0.35 | 0.15–0.86 | .02 |
| Pu Gong Ying | 0.45 | 0.22–0.90 | .02 | 0.36 | 0.18–0.73 | .005 |
| Bai Hua She She Cao | 0.14 | 0.03–0.55 | .005 | 0.15 | 0.04–0.59 | .007 |
| Huang Qi | 0.32 | 0.13–0.77 | .01 | 0.30 | 0.12–0.72 | .007 |
| Dan Shen | 0.53 | 0.24–1.20 | .13 | 0.48 | 0.21–1.07 | .07 |
| Xiang Sha Liu Jun Zi Tang | 0.48 | 0.23–1.03 | .06 | 0.48 | 0.23–1.02 | .06 |
| Ji Xue Teng | 0.49 | 0.22–1.10 | .09 | 0.41 | 0.18–0.93 | .03 |
| Ban Zhi Lian | 0.22 | 0.07–0.69 | .01 | 0.21 | 0.07–0.64 | .007 |
| Gui Pi Tang | 0.42 | 0.19–0.95 | .04 | 0.38 | 0.17–0.87 | .02 |
| Ban Xia | 0.46 | 0.19–1.13 | .09 | 0.51 | 0.21–1.24 | .14 |
- Abbreviations: 95% CI, 95% confidence interval; aHR, adjusted hazards ratio; HR, hazards ratio; TCM, traditional Chinese medicine.
- a The HRs of individual TCM users (≥30 d) were compared with those of nonusers (<30 d).
- b Adjusting for age, congestive heart failure, distant metastases, surgery, hormonal or targeted biologic therapy, and radiotherapy.
DISCUSSION
Clinical studies on the therapeutic effects of TCM in patients with breast cancer remain scant, and to the best of our knowledge the majority are either case discussions or descriptive outcomes of a few patients.9 To our knowledge, this is the first large-scale, nationwide cohort study investigating the association between adjunctive TCM therapy and the survival of patients with advanced breast cancer. Our findings on the effects of TCM on the mortality of patients with breast cancer were similar to those of a population-based study that determined an association between the use of ginseng and survival in these patients.19 By contrast, another study on herbal remedies (TCM and non-TCM) concluded that such remedies did not improve survival in patients with breast cancer.20
The most common herbal formulae used by patients in the current study were similar to another study of patients with all stages of breast cancer in Taiwan.12 According to the concepts of TCM, cancer is a systemic disease formed when the internal functions of the body become imbalanced, which cause toxin accumulation or heat and blood stasis in the body, and this situation eventually leads to the body becoming incapable of resisting external carcinogenic factors.10 The commonly used TCMs aim to treat patient symptoms according to the TCM functional classifications are shown in Table 3. Recent scientific evidence has suggested that Jia Wei Xiao Yao San and Gui Pi Tang can treat insomnia, chronic fatigue, and menopausal syndromes,21-23 whereas other studies observed that Jia Wei Xiao Yao San and Xiang Sha Liu Jun Zi Tang ameliorate functional dyspepsia.24, 25 Apart from the herbal formulae, we also identified the commonly used single herbs. Most of these single herbs were often used to treat cancer in Chinese medicine.26 Recent evidence has demonstrated that the identified TCMs have anticancer effects in vitro27-32 and promising clinical effects in patients.33 Among them, Pu Gong Ying, Bai Hua She She Cao, Dan Shen, Ji Xue Teng, and Ban Zhi Lian have antiproliferative effects on breast cancer cells.27, 28, 30, 31 In addition to anticancer effects, some of the TCMs identified in the current study have been shown to modulate immunity,34 inflammation,35 osteogenesis,36 hematopoiesis,37 and neuroprotection,38, 39 suggesting that in addition to improving patient survival, they also relieve symptoms related to breast cancer or side effects from conventional treatments, and improve quality of life.
Several differences exist between the current study and previous population-based studies. First, the current study data were extracted from a single, large nationwide database with 1 million representative beneficiaries who were randomly selected with minimal selection bias. Second, because TCM is covered by NHI, it was possible to obtain the full TCM prescription records for the patients, which contain data regarding drug names, dose, and days of administration, thus preventing potential recall bias. Moreover, a complete list of TCMs was collected from the database to elucidate the effects of each medicine on patients with breast cancer, which is more comprehensive than merely studying a particular TCM. Third, the TCMs analyzed in the current study were prescribed by board-certified Chinese medicine physicians, and were not TCMs freely acquired by patients. All TCMs covered by NHI must be manufactured by pharmaceutical companies with good manufacturing practice compliant with the regulatory requirements of the central health authority and therefore, the quality and safety of TCMs are ensured. Finally, the study patients were restricted to patients with advanced breast cancer who had similar levels of disease severity, therefore mitigating confounding by indication.40
Several limitations must be considered. First, the NHI program only pays for TCMs prescribed by Chinese medicine physicians, not over-the-counter TCMs. Therefore, the use of TCM may be underestimated. Second, it is likely that TCM users were not fully compliant with prescriptions. However, we were still able to demonstrate the mortality benefit from TCM use. Third, the NHIRD is a claims-based database; therefore, no detailed clinical information regarding cancer staging or biochemical data are recorded. Finally, because the Registry of Catastrophic Illness Database of the NHIRD only documents the date of death and not cause of death, the effect of TCM on breast cancer-specific mortality cannot be analyzed.
The results of the current observational study suggest that adjunctive therapy with TCM may improve the survival of patients with advanced breast cancer who are receiving taxanes. The current study suggests that TCM may be used as an integral element of effective therapy for cancer. Future randomized controlled trials are required to validate these observational findings.
FUNDING SUPPORT
Supported by the Taiwan Food and Drug Administration (DOH99-FDA-61406).
CONFLICT OF INTEREST DISCLOSURES
The authors made no disclosures.
