Sage Journals: Discover world-class research

Abstract

Complementary and alternative medicines are used by up to 48% of lung cancer patients but have seen little formal assessment of survival efficacy. In this 10-year retrospective survival study, the authors investigated Pan-Asian medicine + vitamins (PAM+V) therapy in a consecutive case series of all non-small-cell lung cancer patients (n = 239) presenting at a San Francisco Bay Area Chinese medicine center (Pine Street Clinic). They compared short-term treatment lasting the duration of chemotherapy/radiotherapy with long-term therapy continuing beyond conventional therapy. They also compared PAM+V plus conventional therapy with conventional therapy alone, using concurrent controls from the Kaiser Permanente Northern California and California Cancer Registries. They adjusted for confounding with Kaplan-Meier, Cox regression, and newer methods – propensity score and marginal structural models (MSMs), which when analyzing data from observational studies or clinical practice records can provide results comparable with randomized trials. Long-term use of PAM+V beyond completion of chemotherapy reduced stage IIIB deaths by 83% and stage IV by 72% compared with short-term use only for the duration of chemotherapy. Long-term PAM+V combined with conventional therapy reduced stage IIIA deaths by 46%, stage IIIB by 62%, and stage IV by 69% compared with conventional therapy alone. Survival rates for stage IV patients treated with PAM+V were 82% at 1 year, 68% at 2 years, and 14% at 5 years. PAM+V combined with conventional therapy improved survival in stages IIIA, IIIB, and IV, compared with conventional therapy alone. Prospective trials using PAM+V with conventional therapy for lung cancer patients are justified.

Keywords

lung cancer survival Chinese herbal medicine vitamins propensity score marginal structural models chemotherapy radiotherapy

Background

Lung cancer is the leading cause of cancer death in the United States, with 75% of cases being non-small-cell lung cancer.^1,2 In 1988, when recruitment for the cohort described in the current study began, median survival with inoperable non-small-cell lung cancer was 6 months following etoposide/cisplatin chemotherapy and 8 months with added β- and γ-interferons.³ Systemic therapies for advanced non-small-cell lung cancer have not substantially improved survival. Adjuvant chemotherapy reduces risk of death at 2 years by only 13% compared with surgery alone.⁴ Chemoradiotherapy reduces the risk by 14%⁵; however, adjuvant radiotherapy alone increases it by 21%.^6,7 Platinum-based chemotherapy increased 12-month survival by 5% and tumor response by 62% but with greater toxicity.⁸ In meta-analyses of studies, addition of platinum-based drugs to older chemotherapy regimens improved 12-month survival by only 5% (34% vs 29%; odds ratio [OR] = 1.21, 95% confidence interval [CI] = 1.09, 1.35),⁸ and in individual patient data meta-analyses, 12-month survival after chemotherapy is only 29%.⁹ Gefitinib did not improve survival compared with carboplatin/paclitaxel chemotherapy alone,¹⁰ whereas erlotinib significantly improved survival.¹¹ Improvements in treatment are still needed.

Complementary and alternative medicine (CAM) is a broad field, including physical medicine approaches, such as acupuncture and massage, and pharmacological approaches, such as vitamins and herbal medicine. Herbal medicine is used by approximately 48% of lung cancer patients.¹² Meta-analysis of randomized trials from China suggests that herbal therapies combined with platinum-based chemotherapy increases survival in non-small-cell lung cancer by up to 42%, compared with chemotherapy alone.¹³ However, evidence for survival efficacy in the Western medical literature is sparse. In a retrospective Mayo Clinic cohort study, vitamin use in non-small-cell lung cancer patients was associated with a lower risk of death (hazard ratio [HR] = 0.74; 95% CI = 0.60, 0.91), but that study relied on self-reported and proxy data on vitamins use, did not include herbal medicine, and did not examine survival within disease stage, the most important prognostic variable in non-small-cell lung cancer.¹⁴

Randomized Controlled Trials in CAM Therapies

Although randomized controlled trials are considered the gold standard, they are sometimes unfeasible for researchers and/or undesirable to patients. Less than 5% of cancer patients participate in trials,¹⁵ and little is known about recruitment success in CAM trials.¹⁶ For patients, the common availability of CAM therapies may reduce the acceptability of entering randomized CAM trials. Alternatives to randomized trials exist.¹⁷

Marginal structural model (MSM) methods for causal inference provide near-randomized comparability between treatment groups in observational studies.^18,19 They reduce bias by adjusting for confounding and allow for identification of true causal effects not found through traditional association models such as Cox regression.^20-23 They are a standardization tool, making treatment groups comparable based on the probability of having been treated, given individual characteristics, such as age, gender, and treatment exposure.²⁴ MSMs have not been applied in studies of cancer survival in general²⁵ or Chinese herbal medicine in particular.

The propensity score method for causal inference is another way to create near-randomized comparability between treatment groups in observational studies based on the multivariate probability of receiving treatment.^26,27 It has been used in cancer survival but not in Chinese herbal medicine studies.

Objectives

We conducted this retrospective study to investigate whether Chinese herbal medicine and vitamins improve lung cancer survival and to provide a design foundation and sample size estimates for prospective trials. Our hypotheses were that (1) patients choosing long-term Pan-Asian medicine + vitamins (PAM+V) continuing beyond conclusion of standard therapy would have longer survival than patients using short-term PAM+V lasting only the duration of standard therapy and (2) PAM+V combined with conventional therapy would yield longer patient survival compared with conventional therapy alone.

Methods

Patients

Participants (n = 239) constitute in full a consecutive case series of all patients with non-small-cell lung cancer seen at a Chinese medicine center (Pine Street Clinic, San Anselmo, CA) during 1988-1993 who were also being treated at local oncology centers with surgery, chemotherapy, and/or radiotherapy. In most cases, patients were seen in conjunction with front-line chemotherapy. Patients were diagnosed by physical examination, bronchoscopy, and imaging tests such as X ray and CT scan. They were followed for 10 years or until death or loss to follow-up. We continued in-person visits and/or telephone follow-up for patients continuing at Pine Street Clinic and telephone follow-up for those continuing at other CAM centers. Telephone follow-up for patients following at other centers typically happened every few months and was intended to maintain contact and encourage continuation of therapy. Although we found that this encouragement was beneficial in helping sustain treatment adherence, we did not design the study specifically to study this effect. Those who neither continued at Pine Street nor went to other CAM centers, and therefore for whom no follow-up was possible, were considered lost to follow-up.

Human Subjects Concerns

Prior to initiating this study, we received approval from institutional review boards at the University of California at Berkeley (protocol #2004-6-279), Kaiser Permanente Northern California (protocol #N-04LKush-01-H), the State of California Health and Human Services Agency (protocol #07-08-55), and the Pine Street Foundation (Independent Review Consulting, Corte Madera, California; protocol #EXE04-009-01).

Treatment Rationale

Our treatment design was informed by Chinese herbal strategies which alternate therapeutic phases of attack and rebuilding. Treatment timing is divided into 3 parts, which are chronotherapeutically aimed at complementing the 21-day front-line adjunctive chemotherapy strategy.

Day 1 of the chemotherapy cycle begins part I, which continues through day 4. During these 2 days, the focus is on enhancing circulation, thus enhancing the ability of chemotherapy drugs to circulate as widely as possible through the body to reach tumor cells. Areas of the body that may have suboptimal circulation, for example, where there is tension or a history of surgery, scar tissue, or injuries, are locations where circulation may be inhibited. We use various techniques such as imagery, visualization, exercise, and Qi-Gong to improve circulation. It is also useful during part I to get extra rest and avoid stress.

Days 5 through 10 are called part II, where chemotherapy has killed both cancer cells and healthy cells, which need to be removed. The goal here is to assist the body in reducing lingering chemotherapy-related toxicities by discharging and cleansing, so that these dead cells and remaining toxic chemotherapy metabolites will not be in the way during the next round of therapy. Herbal and supplemental suggestions and lifestyle recommendations around cleansing (home, projects, and personal relationships) also are adapted to this time.

Days 11 through 20 constitute part III, in which the patient is helped to rebuild and enhance the immune system before the next cycle of chemotherapy. Blood counts, physiology, and stamina will begin to normalize, and side effects usually recede. A person’s activities should also be adapted to this, and emphasizing activities that help the individual feel genuinely happy are a premium.

The strategy for patients whose conventional therapy consisted of radiation was to include protocol items for minimizing radiation side effects, minimizing antioxidant therapies during the radiation phase, and resuming antioxidants after radiation. Acupuncture was made available on an as-needed basis, but the primary treatment emphasis was on the herbal/nutraceutical elements. The herbal medicine portion of the PAM+V protocol was administered in the traditional Chinese method of giving patients dry herbs to take home and decoct.

Diet

For all patients, we discussed diet, including basic cancer patient nutrition advice with an emphasis on natural foods, with recommendations to increase fiber and organic foods and improve personal food preparation skills.

Exercise

We advised all patients on exercise, emphasizing basic walking, household and garden chores, and then, if possible, yoga, jogging, gym activities, and training in yoga, Tai Ji, or Qi-Gong.

Data Sources

Data on PAM+V-treated patients were obtained by chart review from Pine Street Clinic (San Anselmo, CA). Survival time in the PAM+V cohort was obtained directly from patient records and/or confirmed by in-person or telephone follow-up with patients’ family members. PAM+V was provided concurrently with chemotherapy/radiation treatment (short-term use). After completion of chemotherapy/radiation, we monitored whether or not the patients elected to continue herbal/vitamin therapy (long-term use). Patients who were lost to follow-up and for whom treatment adherence was, therefore, unknown were analyzed as part of the long-term adherence group, following the intention-to-treat principle.²⁸ Additionally, short-term versus long-term adherence was noted for both patients who after initial consultation here at Pine Street elected to continue at our center and for those who followed up with other CAM centers. Thus, each of these 2 variables was measured separately: location (Pine Street or CAM) and duration (short term or long term).

Data on concurrent external controls diagnosed during the same time span as the Pine Street patients and matched on cancer stage were obtained by database extraction from the California Cancer Registry and Kaiser Permanente Division of Research. As the enrollment period for the Pine Street Clinic cohort began on November 1, 1988, and ended on December 1, 1993, we selected cases from the Kaiser Permanente and California Cancer Registry diagnosed between those dates. For the case selection and exclusion algorithm, see Table 1.

Table 1.

Case Selection and Exclusion Algorithm

	PAM+V	CCR	KPNC	Totals
Original data set	239	85,581	5,617	91,437
Excluded cases
Duplicate record		580	8	588
Not analytic class record		376	14	390
Not San Francisco Bay Area		20,927	32	20,959
Not histologically confirmed		124	77	201
Died within 2 weeks of diagnosis		5,678	265	5,943
Not malignant		70		70
Not NSCLC		12,149	442	12,591
Not adenocarcinoma or squamous		8,408	290	8,698
Unknown AJCC stage		11,030		11,030
In-situ or stage I cases		4,824		4,824
Unknown date of diagnosis	4	4,654		4,658
Unknown date of birth		23		23
Unknown if treated with surgery		392	5	397
Unknown if treated with chemotherapy		534	9	543
Unknown if treated with radiation		3,666	5	3,671
Unknown whether used CAM		293		293
Unknown SEER summary stage			5	5
Stage could not be determined			3,564	3,564
Final data set	235	11,853	901	12,989

Abbreviations: PAM+V, Pan-Asian medicine + vitamins; CCR, California Cancer Registry; KPNC, Kaiser Permanente Northern California; NSCLC, non-small-cell lung cancer; AJCC, American Joint Committee on Cancer; SEER, Surveillance, Epidemiology and End Results.

In Table 2 we report for each cohort the number of participants and their individual and treatment characteristics, such as which conventional treatments they received. In Table 3, we report PAM+V treatment dosage and timing. In Table 4, we report ingredients of Chinese herbal formulas used. The design principles for this whole-systems treatment approach have been described elsewhere.²⁹

Table 2.

Patient Characteristics

	PAM+V	CCR	KPNC	Total
Stage II (number of participants)	33	1202	89	1324
Median age (years)	59	68	66	68
Treated with surgery (%)	100	70	83	71
Treated with chemotherapy (%)	33	11	5	11
Treated with radiation (%)	88	47	53	48
Female (%)	24	39	45	39
Adenocarcinoma (%)	70	51	58	52
Squamous cell carcinoma (%)	30	48	42	48
White (%)	100	87	74	87
Black (%)	0	8	15	8
Asian (%)	0	5	6	5
Hispanic (%)	0	0	6	0
Stage IIIA (number of participants)	75	2598	137	2810
Median age (years)	62	68	65	67
Treated with surgery (%)	20	31	32	31
Treated with chemotherapy (%)	96	23	15	25
Treated with radiation (%)	24	73	77	72
Female (%)	21	40	39	39
Adenocarcinoma (%)	69	48	53	48
Squamous cell carcinoma (%)	31	52	47	51
White (%)	96	89	69	89
Black (%)	1	6	15	6
Asian (%)	3	4	8	4
Hispanic (%)	0	0	8	0
Stage IIIB (number of participants)	84	1974	115	2173
Median age (years)	61	67	65	67
Treated with surgery (%)	0	10	12	10
Treated with chemotherapy (%)	99	31	23	33
Treated with radiation (%)	0	63	71	61
Female (%)	21	39	37	38
Adenocarcinoma (%)	70	52	46	53
Squamous cell carcinoma (%)	30	47	54	47
White (%)	96	88	71	87
Black (%)	4	8	11	8
Asian (%)	0	4	10	4
Hispanic (%)	0	0	7	0
Stage IV (number of participants)	43	6079	560	6682
Median age (years)	61	66	64	66
Treated with surgery (%)	0	7	10	7
Treated with chemotherapy (%)	100	33	21	33
Treated with radiation (%)	0	64	62	64
Female (%)	30	39	43	40
Adenocarcinoma (%)	79	67	72	68
Squamous cell carcinoma (%)	21	32	28	32
White (%)	95	87	73	86
Black (%)	2	8	12	9
Asian (%)	2	5	8	5
Hispanic (%)	0	0	7	1
Total number of participants	235	11 853	901	12 989

Abbreviations: PAM+V, Pan-Asian medicine + vitamins; CCR, California Cancer Registry; KPNC, Kaiser Permanente Northern California.

Table 3.

PAM+V Protocol Timing Concurrent With Chemotherapy, 21-Day Cycle

	Daily Dose	Part I, Days 1-3^a	Part II, Days 4-8	Part III, Days 9-20
Thymus extract	100-500 mg			✓
N-acetyl cysteine	600-1800 mg			✓
Siberian ginseng extract	250-1000 mg		✓	✓
Panax ginseng extract	250-1000 mg		✓	✓
Curcumin	500-1000 mg		✓	✓
Spirulina	1-5 g		✓	✓
Germanium sesquioxide	30-100 mg		✓	✓
Co-enzyme Q-10	90-150 mg		✓	✓
Garlic extract	300-1200 mg		✓	✓
Ginkgo biloba extract	60-180 mg		✓	✓
Green tea beverage	1-5 cups		✓	✓
Grapeseed extract	50-150 mg		✓	✓
Antioxidant combo (Vitamins A, B-complex, B-12, C, D, E, folic acid, selenium, and zinc)			✓	✓
Melatonin	3-20 mg	✓	✓	✓
Acidophilus	1-3 caps	✓	✓	✓
Chinese herbal formula	(See Table 4)	✓	✓	✓
Fish oil	3-10 g	✓	✓	✓
Shark or bovine cartilage	300-1200 mg	✓	✓	✓

Abbreviation: PAM+V, Pan-Asian medicine + vitamins.

Day 1 is defined as the day of chemotherapy infusion. Treatment continues with part III beyond conclusion of chemotherapy.

Table 4.

Chinese Herbal Formula Ingredients

Stage

For external controls obtained from the California Cancer Registry, the American Joint Committee on Cancer (AJCC) staging system was used as the definition for cancer stage for lung cancer patients.³⁰ For external controls obtained from Kaiser Permanente Division of Research, AJCC stage was not available for the study recruitment years (1988-1993), so we used the Surveillance Epidemiology and End Results (SEER) Program Comparative Staging Guide for Cancer^31,32 to manually build TNM staging from data fields for tumor extension, node involvement, and metastasis.

Analysis Methods

We planned to use traditional association models, Kaplan-Meier estimation³³ and Cox proportional hazards regression³⁴ as well as more modern methods—MSMs³⁵ and propensity score²⁶—for causal inference.

Data Analysis

The primary end point was overall survival. Our primary treatment variables of interest were herbal/vitamin therapy (yes/no) and duration of herbal/vitamin therapy (long/short); baseline covariates were stage (II, IIIA, IIIB, or IV), age at diagnosis, sex, race (white/black/Asian), smoking history (ever/never), histology (adenocarcinoma/squamous), and treatment with surgery, radiation, or chemotherapy (yes/no). Time at risk was defined by dates of diagnosis and death (month/year).

We analyzed data in 2 phases: (1) within Pine Street Clinic by comparing patients maintaining long-term Chinese herbal/vitamin therapy with those using only short-term treatment and (2) comparing the Pine Street Clinic cohort with external controls.

For traditional survival analysis, we used Cox proportional hazards regression,³⁴ stratified by stage and adjusted by all covariates, with the stcox program in the Stata statistical package (version 9.2; StataCorp, College Station, TX), and Kaplan-Meier estimation with sts.

To investigate causal inference, we used the MSM approach, in which the Cox proportional hazards model is weighted by the inverse multivariate probability of each patient having received treatment (Inverse Probability Treatment Weighting [IPTW], stabilized). The IPTW balances treatment groups with respect to their probability of having received treatment using logistic regression with the same covariates as our standard Cox regression. We used the bootstrap³⁶ with 1000 repetitions for estimation of the CIs about the HR. We wrote our own program for this analysis, using the Stata statistical software and again stratified by stage.

We also used the propensity score for causal inference, which estimates in a multivariate model the probability of receiving treatment using a person’s individual disease and treatment characteristics, applying the pscore Stata program³⁷ and using the same set of covariates described above. The resulting propensity score was divided into 5 or more levels or strata, and the propensity score then used as a variable in adjusted Cox regression was again stratified by stage. Subclassification of patients into 5 or more levels of the propensity score has been shown to remove 90% of bias.³⁸

Manuscript Preparation

In the drafting of our manuscript, we followed the Strengthening The Reporting of Observational Studies in Epidemiology (STROBE) guidelines.³⁹

Results

Number of Patients

Survival time data suitable for analysis were obtained for 235 of 239 patients from Pine Street Clinic, 11 853 from California Cancer Registry, and 901 from Kaiser (Table 1).

Comparing Long-term With Short-term PAM+V Use

In adjusted Cox regression analysis, survival was significantly better for long-term versus short-term PAM+V in stages IIIB (HR = 0.17; 95% CI = 0.08, 0.36) and IV (HR = 0.28; 95% CI = 0.13, 0.60); there was a large but statistically nonsignificant survival advantage in stages II and IIIA (Table 5).

Table 5.

Lung Cancer Survival With Pan-Asian Medicine and Vitamin Therapy^a

	Stage II		Stage IIIA		Stage IIIB		Stage IV
	HR	CI	HR	CI	HR	CI	HR	CI
PAM+V long- versus short-term
Number of patients (PAM+V long-term)	22		66		71		22
Number of patients (PAM+V short-term)	11		9		13		21
Cox regression, unadjusted	0.77	[0.36, 1.67]	0.82	[0.40, 1.69]	0.21***	[0.11, 0.40]	0.32**	[0.17, 0.62]
Cox regression, adjusted^b	0.73	[0.30, 1.79]	0.49	[0.21, 1.11]	0.17***	[0.08, 0.36]	0.28***	[0.13, 0.60]
Propensity score Cox, unadjusted^c	0.75	[0.30, 1.83]	0.85	[0.40, 1.81]	0.28***	[0.14, 0.56]	0.29**	[0.14, 0.61]
Propensity score Cox, adjusted^b,c	0.82	[0.32, 2.11]	0.50	[0.22, 1.14]	0.17***	[0.08, 0.36]	0.28**	[0.12, 0.61]
MSM Cox regression^d	0.78	[0.39, 1.57]	0.82	[0.17, 3.96]	0.47	[0.09, 2.49]	0.26	[0.18, 3.83]
Long-term PAM+V versus CCR
Number of patients (PAM+V)	22		66		71		22
Number of patients (CCR)	1202		2598		1974		6079
Cox regression, unadjusted	0.74	[0.48, 1.12]	0.51***	[0.40, 0.65]	0.38***	[0.30, 0.49]	0.33***	[0.22, 0.50]
Cox regression, adjusted^e	1.00	[0.64, 1.54]	0.49***	[0.38, 0.63]	0.34***	[0.26, 0.44]	0.36***	[0.23, 0.54]
Propensity score Cox, unadjusted^f	0.91	[0.58, 1.42]	0.54***	[0.41, 0.70]	0.38***	[0.28, 0.50]	0.31***	[0.20, 0.48]
Propensity score Cox, adjusted^e,f	0.92	[0.59, 1.44]	0.51***	[0.39, 0.67]	0.37***	[0.28, 0.50]	0.31***	[0.20, 0.48]
MSM Cox regression^g	0.92	[0.41, 1.43]	0.41***	[0.28, 0.55]	0.44***	[0.31, 0.57]	0.29***	[0.20, 0.38]
Long-term PAM+V versus KPNC
Number of patients (PAM+V)	22		66		71		22
Number of patients (KPNC)	89		137		115		560
Cox regression, unadjusted	0.60*	[0.37, 0.97]	0.39***	[0.29, 0.53]	0.32***	[0.23, 0.44]	0.26***	[0.17, 0.41]
Cox regression, adjusted^e	0.56*	[0.32, 0.99]	0.14***	[0.08, 0.25]	0.23***	[0.13, 0.40]	0.21***	[0.13, 0.34]
Propensity score Cox, unadjusted^f	0.59	[0.32, 1.08]	0.21***	[0.10, 0.42]	0.56*	[0.34, 0.94]	0.09***	[0.04, 0.20]
Propensity score Cox, adjusted^e,f	0.55	[0.30, 1.01]	0.18***	[0.09, 0.35]	0.52*	[0.31, 0.88]	0.09***	[0.04, 0.21]
MSM Cox regression^g	0.66	[0.26, 1.05]	0.40***	[0.30, 0.54]	0.07***	[0.01, 0.65]	0.10**	[0.01, 0.19]
All PAM+V versus CCR
Number of patients (PAM+V)	33		75		84		43
Number of patients (CCR)	1202		2598		1974		6079
Cox regression, unadjusted	0.74	[0.48, 1.12]	0.51***	[0.40, 0.66]	0.38***	[0.30, 0.48]	0.33***	[0.22, 0.50]
Cox regression, adjusted^e	1.00	[0.64, 1.54]	0.49***	[0.38, 0.64]	0.34***	[0.26, 0.44]	0.36***	[0.24, 0.55]
Propensity score Cox, unadjusted^f	0.90	[0.58, 1.40]	0.53***	[0.40, 0.70]^h	0.38***	[0.29, 0.51]	0.32***	[0.21, 0.49]
Propensity score Cox, adjusted^e,f	0.91	[0.59, 1.42]	0.51***	[0.39, 0.67]	0.38***	[0.28, 0.50]	0.31***	[0.20, 0.48]
MSM Cox regression^g	0.91	[0.47, 1.35]	0.42***	[0.28, 0.55]	0.43***	[0.31, 0.56]	0.29***	[0.21, 0.38]
All PAM+V versus KPNC
Number of patients (PAM+V)	33		75		84		43
Number of patients (KPNC)	89		137		115		560
Cox regression, unadjusted	0.64	[0.40, 1.01]	0.42***	[0.31, 0.57]	0.32***	[0.24, 0.44]	0.27***	[0.17, 0.42]
Cox regression, adjusted^e	0.63	[0.37, 1.06]	0.27***	[0.17, 0.43]	0.26***	[0.16, 0.40]	0.25***	[0.16, 0.40]
Propensity score Cox, unadjusted^f	0.64	[0.37, 1.11]	0.40***	[0.25, 0.63]	0.38***	[0.22, 0.65]	0.21***	[0.12, 0.37]
Propensity score Cox, adjusted^e,f	0.61	[0.35, 1.07]	0.29***	[0.18, 0.46]	0.34***	[0.19, 0.59]	0.22***	[0.12, 0.39]
MSM Cox regression^g	0.66	[0.30, 1.03]	0.40***	[0.30, 0.52]	0.14***	[0.07, 0.28]	0.25**	[0.08, 0.41]

Abbreviations: PAM+V, Pan-Asian medicine+vitamins; HR, hazard ratio; CI, confidence interval; MSM, Marginal Structural Model; CCR, California Cancer Registry; KPNC, Kaiser Permanente Northern California.

P < .05; **P < .01; ***P < .001.

Adjusted for age, sex, race, cell type, smoking, surgery, chemotherapy, and radiation.

Propensity score built using age, sex, race, cell type, and smoking and treatment with surgery, chemotherapy, and radiation.

MSM built using age, sex, race, cell type, and smoking and treatment with surgery, chemotherapy, and radiation.

Adjusted for age, sex, race, cell type, surgery, chemotherapy, and radiotherapy.

Propensity score built using age, sex, race, cell type, and treatment with surgery, chemotherapy, and radiation.

Marginal structural models built using age, sex, race, cell type, surgery, chemotherapy, and radiation.

PAM+V compared with CCR: stage IIIA unadjusted propensity score balanced with age, sex, surgery, and radiation.

In adjusted propensity score analysis, survival was significantly better for long-term versus short-term PAM+V in stages IIIB (HR = 0.17; 95% = CI 0.08, 0.36) and IV (HR = 0.28; 95% CI = 0.12, 0.61); there was a modest but statistically nonsignificant survival advantage with long-term PAM+V in stages II and IIIA (Table 5).

In MSM analysis, there was a large but nonsignificant survival advantage to long-term versus short-term PAM+V in all stages (Table 5).

With Kaplan-Meier analysis, survival was substantially better for long-term versus short-term PAM+V in stages IIIB and IV, comparable in stage II, and with a trend toward improved survival in stage IIIA (Figure 1).

Figure 1.

Survival in non-small-cell lung cancer, long-term versus short-term PAM+V and comparison with external controls

Comparing Long-term PAM+V-Treated Patients With Concurrent External Controls From the California Cancer Registry

In adjusted Cox regression, survival was significantly better for long-term PAM+V users versus California Cancer Registry controls in stages IIIA (HR = 0.49; 95% CI = 0.38, 0.63), IIIB (HR = 0.34; 95% CI = 0.26, 0.44), and IV (HR = 0.36; 95% CI = 0.23, 0.54) and was comparable in stage II (HR = 0.92; 95% CI = 0.41, 1.43; Table 5).

In adjusted propensity score analysis, survival was significantly better for long-term PAM+V users versus registry controls in stages IIIA (HR = 0.51; 95% CI = 0.39, 0.67), IIIB (HR = 0.37; 95% CI = 0.28, 0.50) and IV (HR = 0.31; 95% CI = 0.20, 0.48), and there was a nonsignificant trend toward better survival in stage II (Table 5; HR = 0.92; 95% CI = 0.59, 1.44).

In MSM analysis, survival was significantly better for long-term PAM+V users versus registry controls in stage IIIA (HR = 0.41; 95% CI = 0.28, 0.55), stage IIIB (HR = 0.44; 95% CI = 0.31, 0.57), and stage IV (HR = 0.29; 95% CI = 0.20, 0.38), and there was a nonsignificant trend toward better survival in stage II (Table 5; HR = 0.92; 95% CI = 0.41, 1.43).

In Kaplan-Meier analysis, in all stages, long-term PAM+V combined with conventional therapy yielded longer survival compared with conventional therapy alone in registry controls (Figure 1).

Comparing Long-term PAM+V-Treated Patients With Concurrent External Controls From Kaiser Permanente

In adjusted Cox regression, survival was significantly better for long-term PAM+V users versus Kaiser controls in all stages of lung cancer: stage II (HR = 0.56; 95% CI = 0.32, 0.99), stage IIIA (HR = 0.14; 95% CI = 0.08, 0.25), stage IIIB (HR = 0.23; 95% CI = 0.13, 0.40), and stage IV (HR = 0.21; 95% CI = 0.13, 0.34; Table 5).

In adjusted propensity score analysis, survival was significantly better for long-term PAM+V users versus Kaiser controls in stages IIIA (HR = 0.18; 95% CI = 0.09, 0.35), IIIB (HR = 0.52; 95% CI = 0.31, 0.88), and stage IV (HR = 0.09; 95% CI = 0.04, 0.21), and there was a small but nonsignificant trend toward better survival in stage II (HR = 092; 95% CI = 0.59, 1.44; Table 5).

In MSM analysis, survival was significantly better for long-term PAM+V users versus Kaiser controls in stages IIIA (HR = 0.40; 95% CI = 0.30, 0.54), stage IIIB (HR = 0.07; 95% CI = 0.01, 0.65), and stage IV (HR = 0.10; 95% CI = 0.01, 0.19), and there was a large but nonsignificant survival advantage in stage II (HR = 0.66; 95% CI = 0.26, 1.05; Table 5).

In Kaplan-Meier analysis, in all stages, long-term PAM+V combined with conventional therapy yielded longer survival compared with conventional therapy alone in Kaiser controls (Figure 1).

Comparing All PAM+V Users With Concurrent External Controls From the California Cancer Registry

In adjusted Cox regression, survival was significantly better for long-term PAM+V users versus California Cancer Registry controls in stages IIIA (HR = 0.49; 95% CI = 0.38, 0.64), IIIB (HR = 0.34; 95% CI = 0.26, 0.44), and IV (HR = 0.36; 95% CI = 0.24, 0.55) and was comparable in stage II (HR = 1.00; 95% CI = 0.64, 1.54; Table 5).

In adjusted propensity score analysis, survival was significantly better for long-term PAM+V users versus registry controls in stages IIIA (HR = 0.51; 95% CI = 0.39, 0.67), IIIB (HR = 0.37; 95% CI = 0.28, 0.50), and IV (HR = 0.31; 95% CI = 0.20, 0.48), and survival was comparable in stage II (HR = 0.91; 95% CI = 0.59, 1.42; Table 5).

In MSM analysis, survival was significantly better for long-term PAM+V users versus registry controls in stage IIIA (HR = 0.42; 95% CI = 0.28, 0.55), stage IIIB (HR = 0.43; 95% CI = 0.31, 0.56), and stage IV (HR = 0.29; 95% CI = 0.21, 0.38), and there was a small and statistically nonsignificant trend toward better survival in stage II (HR = 0.91; 95% CI = 0.47, 1.35; Table 5).

Comparing All PAM+V-Treated Patients With Concurrent External Controls From Kaiser Permanente

In adjusted Cox regression, survival was significantly better for long-term PAM+V users versus Kaiser controls in stage IIIA (HR = 0.27; 95% CI = 0.17, 0.43), IIIB (HR = 0.26; 95% CI = 0.16, 0.40), and IV (HR = 0.25; 95% CI = 0.16, 0.40; Table 5).

In adjusted propensity score analysis, survival was significantly better for all PAM+V users analyzed together versus Kaiser controls in stage IIIA (HR = 0.29; 95% CI = 0.18, 0.46), stage IIIB (HR = 0.34; 95% CI = 0.19, 0.59), and stage IV (HR = 0.22; 95% CI = 0.12, 0.39), and there was a strong but statistically nonsignificant effect of better survival in stage II (HR = 0.61; 95% CI = 0.35, 1.07; Table 5).

In MSM analysis, survival was significantly better for long-term PAM+V users versus Kaiser controls in stage IIIA (HR = 0.40; 95% CI = 0.30, 0.52), stage IIIB (HR = 0.14; 95% CI = 0.07, 0.28), and stage IV (HR = 0.25; 95% CI = 0.08, 0.41), and there was a large but nonsignificant survival advantage in stage II (HR = 0.66; 95% CI = 0.30, 1.03; Table 5).

Comparing Combinations of PAM+V With Chemotherapy, Surgery, and Radiotherapy

Figure 2 compares Kaplan-Meier survival curves between patients who received PAM+V combined with surgery, PAM+V alone, surgery alone, or neither PAM+V nor surgery stratified by cancer stage and using California Cancer Registry controls. Figure 3 reports all combinations of PAM+V and chemotherapy and Figure 4 all combinations of PAM+V and radiotherapy. Virtually all analyses showed a dose–response effect, with longest survival seen with the full treatment combination of PAM+V with conventional therapy and the least favorable survival response in patients receiving neither treatment, with PAM+V alone or standard therapy alone in the middle (Figures 2 to 4).

Figure 2.

Lung cancer survival, showing all possible treatment combinations of PAM+V and surgery, using California Cancer Registry patients as external controls

Figure 3.

Lung cancer survival, showing all possible treatment combinations of PAM+V and chemotherapy, using California Cancer Registry patients as external controls

Figure 4.

Lung cancer survival, showing all possible treatment combinations of PAM+V and radiotherapy, using California Cancer Registry patients as external controls

Distinguishing the Effect of Treatment From Location of Follow-up

We were also interested in whether there was a “center effect”—that is, survival differences between people continuing long-term PAM+V at Pine Street Clinic and those continuing at other CAM centers. In Kaplan-Meier analysis, we found no apparent difference in survival, with 1 exception of better survival in stage-IV patients continuing at other CAM centers, although this represents only 4 of 43 stage-IV patients (Figure 5).

Figure 5.

Survival in non-small-cell lung cancer, comparing TCM patients initially treated at PSC with those followed up at other CAM centers

CAM Therapy in External Controls

We were fortunate to identify some records within the California Cancer Registry in which clinicians had documented patients’ use of CAM therapies (coded in the parlance of the time as “unproven therapies”) and compared survival with long-term and short-term practitioner-guided PAM+V with those of CAM users from the registry database. In Kaplan-Meier analysis, in stages II, IIIA, IIIB, and IV lung cancer, there was longer survival with long-term practitioner-guided PAM+V therapy compared with CAM use in the general population (Figure 6). In patients with stages IIIA, IIIB, and IV cancer, there was no difference in survival between short-term PAM+V users within the Pine Street cohort and CAM users in the general population, further reinforcing the benefit of long-term maintenance of adjunctive complementary therapies (although these findings on CAM use in the general population are prone to error because of misclassification bias arising from the way in which CAM use data were gathered by the California Cancer Registry (Figure 6).

Figure 6.

Survival in non-small-cell lung cancer, comparing patients treated with long-term and short-term PAM+V with CAM users and nonusers from within the CCR database

Validity of External Controls

To confirm the validity of external controls, we compared survival rates from our analysis among California Cancer Registry and Kaiser controls with those reported by other authors also using SEER data and found that they were comparable (Table 6). For example, in stage IIIB and IV patients treated with chemotherapy, median survival in California Cancer Registry controls was 6.9 months and in Kaiser controls 6.1 months, compared with 6.8 months in other studies also using SEER data during the same time period.⁴⁰

Table 6.

Absolute Survival Rates for Examining the Validity of External Controls

Stage	Mortality Statistic	Subpopulation	Our Analysis: CCR	Our Analysis: KPNC	Reference Comparison	Data Source
II	Median survival	Unresected, treated with radiation	13.3 Months	10.5 Months	14 Months	SEER⁴¹
IIIA	5-Year survival	Surgically resected, age <79 years	23%	17%	24%	SEER⁴²
IIIB and IV	Median survival	Treated with chemotherapy	6.9 Months	6.1 Months	6.8 Months	SEER⁴⁰
IIIB and IV	Median survival	Treated with chemotherapy	6.9 Months	6.1 Months	6.8 Months	Phase III trial⁴³

Abbreviations: CCR, California Cancer Registry; KPNC, Kaiser Permanente Northern California; SEER, Surveillance, Epidemiology and End Results, a program established in 1973 by the National Cancer Institute that collects information on cancer incidence, prevalence and survival from specific geographic areas representing 26% of the US population.

Crude Survival Rates: Stage II

Survival at 1 year was 95% in the long-term PAM+V group, 100% in the short-term PAM+V group, 64% in the Kaiser controls, and 67% in California Cancer Registry controls. Survival at 2 years was 93% in the long-term PAM+V group, 70% in the short-term PAM+V group, 50% in Kaiser controls, and 47% in California Cancer Registry controls. Survival at 5 years was 36% in both long-term and short-term PAM+V groups, 13% in Kaiser controls, and 22% in California Cancer Registry controls (Table 7).

Table 7.

Survival Rates at 1, 2, and 5 Years

	Long-term PAM+V	Short-term PAM+V	Kaiser Permanente	California Cancer Registry
Stage II
1 Year	95%	100%	64%	67%
2 Year	77%	82%	37%	44%
5 Year	36%	36%	13%	22%
Stage IIIA
1 Year	93%	70%	50%	47%
2 Year	83%	44%	23%	25%
5 Year	32%	22%	8%	11%
Stage IIIB
1 Year	89%	23%	34%	29%
2 Year	72%	15%	11%	12%
5 Year	24%	0%	5%	4%
Stage IV
1 Year	82%	24%	16%	17%
2 Year	60%	10%	4%	6%
5 Year	14%	5%	1%	2%

Crude Survival Rates: Stage IIIA

Survival at 1 year was 93% in the long-term PAM+V group, 70% in the short-term PAM+V group, 50% in Kaiser controls, and 47% in California Cancer Registry controls. Survival at 2 years was 83% in the long-term PAM+V group, 44% in the short-term PAM+V group, 23% in Kaiser controls and 25% in California Cancer Registry controls. Survival at 5 years was 32% in the long-term PAM+V group, 22% in the short-term PAM+V group, 8% in Kaiser controls, and 11% in California Cancer Registry controls (Table 7).

Crude Survival Rates: Stage IIIB

Survival at 1 year was 89% in the long-term PAM+V group, 23% in the short-term PAM+V group, 34% in Kaiser controls, and 29% in California Cancer Registry controls. Survival at 2 years was 72% in the long-term PAM+V group, 15% in the short-term PAM+V group, 11% in Kaiser controls, and 12% in California Cancer Registry controls. Survival at 5 years was 24% in the long-term PAM+V group, 0% in the short-term PAM+V group, 5% in Kaiser controls, and 4% in California Cancer Registry controls (Table 7).

Crude Survival Rates: Stage IV

Survival at 1 year was 82% in the long-term PAM+V group, 24% in the short-term PAM+V group, 16% in Kaiser controls, and 17% in California Cancer Registry controls. Survival at 2 years was 60% in the long-term PAM+V group, 10% in the short-term PAM+V group, 4% in Kaiser controls, and 6% in California Cancer Registry controls. Survival at 5 years was 14% in the long-term PAM+V group, 5% in the short-term PAM+V group, 1% in Kaiser controls, and 2% in California Cancer Registry controls (Table 7).

Discussion

Principal Findings

In our retrospective survival analysis for non-small-cell lung cancer, we found a substantial and statistically significant survival advantage to combining PAM+V with conventional therapy compared with treatment with conventional therapy alone; this was strongest in advanced lung cancer—stages IIIA, IIIB, and IV. We also found significant benefits of long-term PAM+V use compared with short-term use discontinued after adjuvant chemotherapy and radiotherapy. MSM and propensity score analyses found results usually consistent with multivariate Cox regression analysis. Propensity score analysis had a stronger ability to detect a survival benefit, particularly in subanalyses with small sample sizes.

Strengths of the Study

To our knowledge, this is the first use of causal inference using the MSM approach in cancer survival or in any CAM therapy. It is also the first use of the propensity score in any Chinese herbal therapy.

We intentionally used specific herbal and vitamin products commonly available in the commercial markets and free of proprietary constraints. In this way, we hoped to maximize the accessibility and applicability of our treatment approach for clinicians and researchers.

We found consistency between the different multivariate analyses methods used: traditional Cox regression, MSM Cox regression, and propensity score Cox regression. Propensity score analysis was best able to detect a survival advantage, followed by MSMs.

Lead time (the delay between diagnosis and initiation of PAM+V treatment) averaged between 22 days and 28 days. This is comparable with what is known about delays in delivery of care to those with lung cancer in conventional oncology practice: delays attributable to the patient average 18 days, and delays attributable to the health system average 62 days.⁴⁴ To strengthen our inference, we conducted a sensitivity analysis in which we dropped all PAM+V-treated patients with lead time greater than 60 days. We also conducted a conservative analysis where we excluded all KPNC external controls who survived for less than 60 days. In both these sensitivity analyses, we found differences of only a few percentage points in the HRs, which did not change our inferences.

Patients’ use of imagery, visualization, exercise, and Qi-Gong was an integral part of the treatment design, and they were encouraged to learn and practice these approaches as much as possible; these are the kinds of approaches studied in whole systems research, where the emphasis is on the net effect of the holistic protocol design. Teasing apart the relative efficacy of each individual PAM+V component could be pursued in a future prospective study designed to answer this question. Holistic protocols like PAM+V can also be studied using an approach called pragmatic trials, which evaluate a therapy as it is used in normal practice (as compared with the more fixed and constrained approach used in most clinical trials).⁴⁵

Limitations of the Study

These are observational data, retrospectively gathered from medical records and cancer registry databases. There may have been unmeasured differences between patients choosing PAM+V versus those who did not, which could have additionally contributed to differences in survival. We did not have data that would allow us to control for possible confounding by socioeconomic status, which is associated with increased use of CAM,⁴⁶ better access to conventional cancer therapy,⁴⁷ and increased cancer survival.⁴⁸

In the California Cancer Registry and Kaiser, we did not have available data on smoking, which is an important prognostic factor in lung cancer survival. Although some studies have shown that smoking history confers less favorable survival in people diagnosed with lung cancer,^49,50 others have shown no significant differences in survival,⁵¹ particularly in people with advanced disease.⁵² We did, however, have smoking history data available for those in the PAM+V cohort. This allowed us to include smoking as a variable in all our multivariate survival analyses comparing long-term versus short-term PAM+V treatment.

We did not have prospectively gathered data available for analysis and were therefore unable to account for time-dependent confounding, which is often present in longitudinal studies and capably handled by MSM analysis. Early papers on the MSMs stated the assumption of no unobserved confounders, although this assumption is unverifiable and has recently been called into question.^53,54

We did not have data on CAM use by external controls in the Kaiser data. Nevertheless, even if the proportion of CAM use among Kaiser members was as high as the 26% to 30% reported in the literature,⁵⁵ we anticipate that this would have biased our results toward the null and that the true survival benefit of PAM+V use may even be stronger than we estimated. To strengthen our inference, we also conducted a conservative analysis where we excluded all KPNC external controls who survived less than 2 months and found only a difference in the HR of a few percentage points (which caused no change in our inference).

The existing and newly developed conventional therapies for lung cancer, which were in use during the time span of data collection for this study, are known to have significant toxicities.^56-58 Enhancing patient compliance with and tolerance of conventional therapies is an often-reported goal of many CAM treatment programs. We did not include that outcome as a part of the current study. Nevertheless, we are starting to see outcomes data on the contributions of Chinese herbal therapy to chemotherapy compliance in a meta-analysis we are currently conducting.

Causal Inference Methods Used in Our Study

This study uses 2 newer analysis methods for what is called causal inference: the propensity score and the MSM. Causal inference attempts to look more closely into the data being analyzed to uncover causal relationships. This is not always possible with traditional statistical methods, which can usually conclude only that the observed association between 2 variables is not a result of chance and cannot identify which variable causes the other.

Both the propensity score and MSM methods begin by using multivariable analysis to identify for each individual in the study, based on their personal and treatment history, that that individual chose one treatment or another. The propensity score and MSM estimate that score with similar but slightly different statistical approaches. The propensity score is used as a variable in the Cox regression. The MSM is similarly used to weight the Cox regression.

In practice, both methods of causal inference will balance the treatment groups, so that patients in each group are comparable with respect to confounding variables. The primary practical difference between the MSM and propensity score methods is that the MSM can adequately handle the effects of treatment variables that vary over time, based on the outcomes of treatment. In the case of our study, however, this difference is not relevant because our study was of the point treatment design, where variables are measured only at baseline.

Possible Mechanisms

The design of the treatment protocol used in this study was based on treatment strategies from traditional Chinese medicine, combining a large number of different herbal and vitamin substances into 1 integrated treatment strategy. The timing for each protocol item described below is listed in Table 3.

Several plausible mechanisms for these findings may be considered, based on what is known about the specific components of this combination treatment approach with respect to various effects.

General cancer cell cycle effects

Green tea has been shown to reduce the risk of lung cancer in several case control studies,⁵⁹ notably in smokers.⁶⁰ N-acetyl cysteine inhibits tumor growth^61-63 and metastasis.⁶⁴ Curcumin is extracted from the culinary spice turmeric (Curcuma longa) and inhibits cell invasion and metastasis,⁶⁵ inhibits NF-κB and COX-2 in human non-small-cell lung cancer cells,⁶⁶ and has antiproliferative⁶⁷ and apoptosis-stimulating functions.⁶⁸ Garlic stimulates apoptosis in non-small-cell lung cancer cells.⁶⁹ Fish oil possesses anti-inflammatory and antitumor activity.

Chemotherapy-related effects

Green tea enhances the antitumor efficacy of anthracycline, cisplatin, and irinotecan chemotherapy.⁷⁰ Green tea also contains the polyphenol epigallocatechin-3-gallate, which inhibits telomerase and induces apoptosis in drug-resistant lung cancer cells.⁷¹

Reduction of chemotherapy toxicity and side effects

The green tea plant contains l-theanine, which has neuroprotective⁷² and cognition-enhancing properties.⁷³ N-acetyl cysteine is a liver protective antioxidant^74-76 that reduces chemotherapy toxicity.^77-79 Co-enzyme Q-10 prevents chemotherapy-related cardiotoxicity.⁸⁰ Lactobacillus reduces the frequency of severe diarrhea and abdominal discomfort from chemotherapy.⁸¹

Reduction of radiation toxicity and side effects

Components found within the ginseng root protect against DNA damage from radiation exposure,⁸² potentiate the therapeutic effect of radiation on lung cancer cells,⁸³ and reduce radiation-induced enteritis.⁸⁴ Cordyceps protects against radiation-induced bone marrow and intestinal injuries,⁸⁵ accelerates leukocyte recovery, stimulates immune lymphocyte proliferation, and improves animal survival after radiation exposure.⁸⁶ The Chinese herbal medicines Radix angelicae and Radix paeoniae protect intestinal cells from irradiation-induced damage.⁸⁷ Furthermore, Radix angelicae inhibits radiation-induced pulmonary fibrosis, with the mechanism being downregulation of proinflammatory cytokine expression.⁸⁸ Radix atractylodes and Poria cocos reduce radiation injury to healthy cells.⁸⁹ Radix rehmannia and Radix ligustici reduced radiation pneumonitis in a study of 46 individuals with non-small-cell lung cancer (incidence of radiation pneumonitis was 26% in controls and 10% in the herbal group).⁹⁰ Gingko biloba and Panax ginseng protect against radiation-induced lethality, lipid peroxidation, and DNA damage.⁸² Docosahexaenoic acid, an n-3 polyunsaturated fatty acid found in fish oil, in combination with radiation significantly increased oxidative stress and cell death in a lung cancer cell line when compared with either treatment alone.⁹¹ It accomplishes this by increasing heme oxygenase levels in lung endothelial and fibroblast cells and by blocking radiation-induced reactive oxygen species. Curcumin, an active component of the curry seasoning turmeric, reduced radiation-induced pulmonary fibrosis in mice and increased their survival, without reducing the tumor cell kill rate of radiation therapy.⁹² Caffeic acid phenylethyl ester, a component of propolis, reduced radiation-induced lung injury in a rat model through its antioxidant effect.⁹³ Vitamin E provided immediately after radiation therapy in a rat model protected against radiation-induced pulmonary fibrosis.⁹⁴ Melatonin reduces radiation-induced oxidative organ damage in rats by increasing malondialdehyde levels, myeloperoxidase activity, and glutathione levels.⁹⁵

More general systemic effects such as immune system modulation

N-acetyl cysteine enhances the functionality of peripheral blood mononuclear cells.⁹⁶ Thymic protein is obtained from a cloned cell line of thymic epithelial origin and induces T-cell maturation.⁹⁷ In lung cancer patients, fish oil increases appetite and body weight, reduces fatigue, improves muscle strength and lowers C-reactive protein levels.⁹⁸ Melatonin is a sleep-regulating hormone shown to improve survival^99,100 and tumor response¹⁰¹ in lung cancer patients when combined with conventional therapies.

An immune-enhancing Chinese herbal formula was also used, which combines herbal medicines traditionally recognized as tonifying blood yin (nourishes bone marrow), tonifying Spleen yin/qi (boosts cellular immunity), harmonizing Liver qi (balances liver detoxification pathways), and tonifying Kidney yang (improves dispositional optimism). The ingredients are shown in Table 3.

Summary and Future Directions

Although previous work has examined the survival benefit of Chinese herbal therapies in combination with standard chemotherapy for lung cancer,¹³ there is little available evidence for whole-systems combination CAM therapies. In this retrospective survival analysis comparing patients treated with long-term versus short-term PAM+V therapy, there was a substantial, statistically significant, survival advantage in stages IIIA, IIIB, and IV and a modest but nonsignificant advantage in stage II. PAM+V combined with standard therapy, compared with standard therapy alone, achieved significantly better survival in stages IIIA, IIIB, and IV disease and conferred a modest but nonsignificant advantage in stage II; these results were stronger when comparing long-term users of PAM+V with external controls.

It is important to emphasize in both analyses that the survival advantages reflect the benefit of using PAM+V combined with, rather than instead of, conventional therapy. Our results suggest that prospective trials for advanced non-small-cell lung cancer patients using PAM+V and vitamin therapy, in a whole-systems approach combined with conventional therapies, are justified.

Footnotes

Acknowledgements

To my instructors for providing guidance, to Dr Ly Suhuai and Michael Broffman for inspiration, to Louise Estupinian for assistance in records acquisition, to the kind members of Statalist for their advice, and to my family for their loving support [MM].

Author contributions are as follows: Conception and design; M. F. McCulloch, M. D. Broffman, J. M. Colford Jr. Administrative support; M. D. Broffman, L. Kushi, J. Gao, J. M. Colford Jr. Provision of study materials or patients; M. F. McCulloch, M. D. Broffman, L. Kushi. Collection and assembly of data; M. F. McCulloch, M. D. Broffman, J. Gao. Data analysis and interpretation; M. F. McCulloch, M. van der Laan, A. Hubbard, A. Kramer, J. M. Colford Jr. Manuscript writing; M. F. McCulloch, J. M. Colford Jr. The collection of cancer incidence data used in this study was supported by the California Department of Public Health as part of the statewide cancer reporting program mandated by California Health and Safety Code Section 103885; the National Cancer Institute’s Surveillance, Epidemiology and End Results Program under contract N01-PC-35136 awarded to the Northern California Cancer Center, contract N01-PC-35139 awarded to the University of Southern California, and contract N02-PC-15105 awarded to the Public Health Institute; and the Centers for Disease Control and Prevention’s National Program of Cancer Registries, under agreement #U55/CCR921930-02 awarded to the Public Health Institute. The ideas and opinions expressed herein are those of the author(s) and endorsement by the State of California, Department of Public Health, the National Cancer Institute, and the Centers for Disease Control and Prevention or their contractors and subcontractors is not intended nor should it be inferred.

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by the Pine Street Foundation, a University of California at Berkeley Epidemiology Block Grant Award, and a University of California at Berkeley Research Fellowship.

References

Jemal

Siegel

Ward

. Cancer statistics, 2006. CA Cancer J Clin. 2006;56:106-130.

Herbst

Heymach

Lippman

. Lung cancer. N Engl J Med. 2008;359:1367-1380.

Schiller

Storer

Dreicer

. Randomized phase II-III trial of combination beta and gamma interferons and etoposide and cisplatin in inoperable non-small cell cancer of the lung. J Natl Cancer Inst. 1989;81:1739-1743.

Non-small Cell Lung Cancer Collaborative Group. Chemotherapy for non-small cell lung cancer. Cochrane Database Syst Rev. 2004;(4):CD002139.

Rowell

O’Rourke

. Concurrent chemoradiotherapy in non-small cell lung cancer. Cochrane Database Syst Rev. 2004;(4):CD002140.

PORT Meta-analysis Trialists Group. Postoperative radiotherapy for non-small cell lung cancer. Cochrane Database Syst Rev. 2004;(4):CD002142.

Burdett

Stewart

. Postoperative radiotherapy in non-small-cell lung cancer: update of an individual patient data meta-analysis. Lung Cancer. 2005;47:81-3.

D’Addario

Pintilie

Leighl

. Platinum-based versus non-platinum-based chemotherapy in advanced non-small-cell lung cancer: a meta-analysis of the published literature. J Clin Oncol. 2005;23:2926-2936.

NSCLC Meta-Analyses Collaborative Group. Chemotherapy in addition to supportive care improves survival in advanced non-small-cell lung cancer: a systematic review and meta-analysis of individual patient data from 16 randomized controlled trials. J Clin Oncol. 2008;26: 4617-4625.

10.

Herbst

Giaccone

Schiller

. Gefitinib in combination with paclitaxel and carboplatin in advanced non-small-cell lung cancer: a phase III trial: INTACT 2. J Clin Oncol. 2004;22:785-794.

11.

Murdoch

Sager

. Will targeted therapy hold its promise? An evidence-based review. Curr Opin Oncol. 2008;20: 104-111.

12.

Molassiotis

Panteli

Patiraki

. Complementary and alternative medicine use in lung cancer patients in eight European countries. Complement Ther Clin Pract. 2006;12:34-39.

13.

McCulloch

See

Shu

. Astragalus-based Chinese herbs and platinum-based chemotherapy for advanced non-small-cell lung cancer: meta-analysis of randomized trials. J Clin Oncol. 2006;24:419-430.

14.

Jatoi

Williams

Nichols

. Is voluntary vitamin and mineral supplementation associated with better outcome in non-small cell lung cancer patients? Results from the Mayo Clinic lung cancer cohort. Lung Cancer. 2005;49:77-84.

15.

Murthy

Krumholz

Gross

. Participation in cancer clinical trials: race-, sex-, and age-based disparities. JAMA.2004;291:2720-2726.

16.

Smith

Coyle

. Recruitment and implementation strategies in randomised controlled trials of acupuncture and herbal medicine in women’s health. Complement Ther Med. 2006;14:81-86.

17.

West

Duan

Pequegnat

. Alternatives to the randomized controlled trial. Am J Public Health. 2008;98: 1359-1366.

18.

Cole

Hernan

Margolick

. Marginal structural models for estimating the effect of highly active antiretroviral therapy initiation on CD4 cell count. Am J Epidemiol. 2005;162:471-478.

19.

Petersen

Wang

van der Laan

. Assessing the effectiveness of antiretroviral adherence interventions: using marginal structural models to replicate the findings of randomized controlled trials. J Acquir Immune Defic Syndr. 2006;43(suppl 1):S96-S103.

20.

Hernan

Brumback

Robins

. Marginal structural models to estimate the joint causal effect of nonrandomized treatments. J Am Stat Assoc. 2001;96:440-448.

21.

Robins

Finkelstein

. Correcting for noncompliance and dependent censoring in an AIDS Clinical Trial with inverse probability of censoring weighted (IPCW) log-rank tests. Biometrics. 2000;56:779-788.

22.

Mortimer

Neugebauer

van der Laan

. An application of model-fitting procedures for marginal structural models. Am J Epidemiol. 2005;162:382-388.

23.

Robins

Hernan

Brumback

. Marginal structural models and causal inference in epidemiology. Epidemiology. 2000;11:550-560.

24.

Sato

Matsuyama

. Marginal structural models as a tool for standardization. Epidemiology. 2003;14:680-686.

25.

Yamaguchi

Ohashi

. Adjusting for differential proportions of second-line treatment in cancer clinical trials. Part I: structural nested models and marginal structural models to test and estimate treatment arm effects. Stat Med. 2004;23:1991-2003.

26.

Rosenbaum

Rubin

. The central role of the propensity score in observational studies for causal effects. Biometrika. 1983;70:41-55.

27.

Rosenbaum

Rubin

. Reducing bias in observational studies using subclassification on the propensity score. J Am Stat Assoc. 1984;79:516-524.

28.

Newell

. Intention-to-treat analysis: implications for quantitative and qualitative research. Int J Epidemiol. 1992;21: 837-841.

29.

Broffman

McCulloch

See

. Magic fusion: the meaning of integration in medicine. In: Preedy

RRWVR

, ed. Botanical Medicine in Clinical Practice. Wallingford, UK: CABI; 2008, 366-76.

30.

Grondin

Liptay

. Current concepts in the staging of non-small cell lung cancer. Surg Oncol. 2002;11:181-190.

31.

National Cancer Institute. SEER Program: Comparative Staging Guide for Cancer. Bethesda, MD: National Cancer Institute; 1993. NIH Publication 93-3640.

32.

Johnson

Adamo

, eds. SEER Program Coding and Staging Manual 2007. Bethesda, MD: National Cancer Institute; 2007.

33.

Kaplan

Meier

. Non-parametric estimation for incomplete observations. J Am Stat Assoc. 1958;53:457-481.

34.

Cox

. Regression models and life tables (with discussion). J R Stat Soc Series B Stat Methodol. 1972;34:187-220.

35.

Robins

. Marginal structural models. In: Proceedings of the Section on Bayesian Statistical Science. Alexandria, VA: American Statistical Association; 1997:1-10.

36.

Efron

. Bootstrap methods: another look at the jackknife. Ann Stat. 1979;7:1-26.

37.

Becker

Ichino

. Estimation of average treatment effects based on propensity scores. Stata J. 2002;2:358-377.

38.

Rubin

. Estimating causal effects from large data sets using propensity scores. Ann Intern Med. 1997;127(8, pt 2): 757-763.

39.

Vandenbroucke

von Elm

Altman

. Strengthening the reporting of observational studies in epidemiology (STROBE): explanation and elaboration. PLoS Med. 2007; 4:e297.

40.

Breathnach

Freidlin

Conley

. Twenty-two years of phase III trials for patients with advanced non-small-cell lung cancer: sobering results. J Clin Oncol. 2001;19:1734-1742.

41.

Wisnivesky

Bonomi

Henschke

. Radiation therapy for the treatment of unresected stage I-II non-small cell lung cancer. Chest. 2005;128:1461-1467.

42.

Ravdin

Davis

. Prognosis of patients with resected non-small cell lung cancer: impact of clinical and pathologic variables. Lung Cancer. 2006;52:207-212.

43.

Wozniak

Crowley

Balcerzak

. Randomized trial comparing cisplatin with cisplatin plus vinorelbine in the treatment of advanced non-small-cell lung cancer: a Southwest Oncology Group study. J Clin Oncol. 1998;16:2459-2465.

44.

Valdes

Garcia

Perez

. Length of diagnostic delay in patients with non-small-cell lung cancer. MEDICC Rev. 2010;12:29-32.

45.

Macpherson

. Pragmatic clinical trials. Complement Ther Med. 2004;12:136-140.

46.

Helyer

Chin

Chui

. The use of complementary and alternative medicines among patients with locally advanced breast cancer: a descriptive study. BMC Cancer. 2006;6:39.

47.

Dejardin

Bouvier

Faivre

. Access to care, socioeconomic deprivation and colon cancer survival. Aliment Pharmacol Ther. 2008;27:940-949.

48.

Zell

Ziogas

. Low socioeconomic status is a poor prognostic factor for survival in stage I nonsmall cell lung cancer and is independent of surgical treatment, race, and marital status. Cancer. 2008;112:2011-2020.

49.

Goodman

Kolonel

Wilkens

. Smoking history and survival among lung cancer patients. Cancer Causes Control. 1990;1:155-163.

50.

Tammemagi

Neslund-Dudas

Simoff

. In lung cancer patients, age, race-ethnicity, gender and smoking predict adverse comorbidity, which in turn predicts treatment and survival. J Clin Epidemiol. 2004;57:597-609.

51.

Toh

Wong

Lim

. The impact of smoking status on the behavior and survival outcome of patients with advanced non-small cell lung cancer: a retrospective analysis. Chest. 2004;126:1750-1756.

52.

Fox

Rosenzweig

Ostroff

. The effect of smoking status on survival following radiation therapy for non-small cell lung cancer. Lung Cancer. 2004;44:287-293.

53.

Lefebvre

Delaney

Platt

. Impact of mis-specification of the treatment model on estimates from a marginal structural model. Stat Med. 2008;27:3629-3642.

54.

Cole

Hernan

. Constructing inverse probability weights for marginal structural models. Am J Epidemiol. 2008;168:656-664.

55.

Gordon

Lin

. Use of complementary and alternative medicine by the adult membership of a large northern California health maintenance organization, 1999. J Ambul Care Manage. 2004;27:12-24.

56.

Choi

Kanarek

. Toxicity of thoracic radiotherapy on pulmonary function in lung cancer. Lung Cancer. 1994;10 Suppl 1: S219-230.

57.

Van Houtte

Danhier

. Toxicity of combined radiation and chemotherapy in non-small cell lung cancer. Lung Cancer. 1994; 10 Suppl 1: S271-280.

58.

Brundage

Davidson

. Trading treatment toxicity for survival in locally advanced non-small cell lung cancer. Journal of Clinical Oncology: official journal of the American Society of Clinical Oncology. 1997;15(1): 330-340.

59.

Zhong

Goldberg

Gao

. A population-based case-control study of lung cancer and green tea consumption among women living in Shanghai, China. Epidemiology. 2001;12:695-700.

60.

Liang

Binns

Jian

. Does the consumption of green tea reduce the risk of lung cancer among smokers? Evid Based Complement Alternat Med.2007;4:17-22.

61.

Hecht

Upadhyaya

Wang

. Inhibition of lung tumorigenesis in A/J mice by N-acetyl-S-(N-2-phenethylthiocarbamoyl)-l-cysteine and myo-inositol, individually and in combination. Carcinogenesis. 2002;23: 1455-1461.

62.

Qanungo

Wang

Nieminen

. N-Acetyl-l-cysteine enhances apoptosis through inhibition of nuclear factor-kappaB in hypoxic murine embryonic fibroblasts. J Biol Chem. 2004;279:50455-50464.

63.

Simkeviciene

Straukas

Uleckiene

. N-acetyl-l-cysteine and 2-amino-2-thiiazoline N-acetyl-l-cysteinate as a possible cancer chemopreventive agents in murine models. Acta Biol Hung. 2002;53:293-298.

64.

Goldman

Peled

Shinitzky

. Effective elimination of lung metastases induced by tumor cells treated with hydrostatic pressure and N-acetyl-l-cysteine. Cancer Res. 2000;60:350-358.

65.

Chen

Lee

Huang

. Curcumin inhibits lung cancer cell invasion and metastasis through the tumor suppressor HLJ1. Cancer Res. 2008;68:7428-7438.

66.

Lee

Jung

. Curcumin inhibits interferon-alpha induced NF-kappaB and COX-2 in human A549 non-small cell lung cancer cells. Biochem Biophys Res Commun. 2005;334:313-318.

67.

Zhang

. Research of anti-proliferation of curcumin on A549 human lung cancer cells and its mechanism [in Chinese]. Zhong Yao Cai. 2004;27:923-927.

68.

Radhakrishna Pillai

Srivastava

Hassanein

. Induction of apoptosis in human lung cancer cells by curcumin. Cancer Lett. 2004;208:163-170.

69.

Hong

Ham

Choi

. Effects of allyl sulfur compounds and garlic extract on the expression of Bcl-2, Bax, and p53 in non small cell lung cancer cell lines. Exp Mol Med. 2000;32:127-134.

70.

Sugiyama

Sadzuka

. Theanine and glutamate transporter inhibitors enhance the antitumor efficacy of chemotherapeutic agents. Biochim Biophys Acta. 2003;1653: 47-59.

71.

Sadava

Whitlock

Kane

. The green tea polyphenol, epigallocatechin-3-gallate inhibits telomerase and induces apoptosis in drug-resistant lung cancer cells. Biochem Biophys Res Commun. 2007;360:233-237.

72.

Kakuda

. Neuroprotective effects of the green tea components theanine and catechins. Biol Pharm Bull. 2002;25: 1513-1518.

73.

Nathan

Gray

. The neuropharmacology of l-theanine(N-ethyl-l-glutamine): a possible neuroprotective and cognitive enhancing agent. J Herb Pharmacother. 2006;6:21-30.

74.

Ioannides

Hall

Mulder

. A comparison of the protective effects of N-acetyl-cysteine and S-carboxymethylcysteine against paracetamol-induced hepatotoxicity. Toxicology. 1983;28:313-321.

75.

Sprague

Elfarra

. Protection of rats against 3-butene-1,2-diol-induced hepatotoxicity and hypoglycemia by N-acetyl-l-cysteine. Toxicol Appl Pharmacol. 2005;207:266-274.

76.

Varma

Aruna

Rukkumani

. Alcohol and thermally oxidized PUFA induced oxidative stress: role of N-acetyl cysteine. Ital J Biochem. 2004;53:10-15.

77.

Berend

. Inhibition of bleomycin lung toxicity by N-acetyl cysteine in the rat. Pathology. 1985;17:108-110.

78.

Gon

Hashimoto

Nakayama

. N-acetyl-l-cysteine inhibits bleomycin-induced interleukin-8 secretion by bronchial epithelial cells. Respirology. 2000;5:309-313.

79.

Trizna

Schantz

Hsu

. Effects of N-acetyl-l-cysteine and ascorbic acid on mutagen-induced chromosomal sensitivity in patients with head and neck cancers. Am J Surg. 1991;162:294-298.

80.

Takimoto

Sakurai

Kodama

. Protective effect of CoQ 10 administration on cardial toxicity in FAC therapy [in Japanese]. Gan To Kagaku Ryoho. 1982;9:116-121.

81.

Osterlund

Ruotsalainen

Korpela

. Lactobacillus supplementation for diarrhoea related to chemotherapy of colorectal cancer: a randomised study. Br J Cancer. 2007;97:1028-1034.

82.

Jagetia

. Radioprotective potential of plants and herbs against the effects of ionizing radiation. J Clin Biochem Nutr. 2007;40:74-81.

83.

Chae

Kang

Chang

. Effect of compound K, a metabolite of ginseng saponin, combined with gamma-ray radiation in human lung cancer cells in vitro and in vivo. J Agric Food Chem. 2009;57:5777-5782.

84.

Takeda

Kamiura

Kimura

. Effectiveness of the herbal medicine daikenchuto for radiation-induced enteritis. J Altern Complement Med. 2008;14:753-755.

85.

Liu

Wang

Tsai

. Protection against radiation-induced bone marrow and intestinal injuries by Cordyceps sinensis, a Chinese herbal medicine. Radiat Res. 2006;166:900-907.

86.

Xun

Shen

. Radiation mitigation effect of cultured mushroom fungus Hirsutella sinensis (CorImmune) isolated from a Chinese/Tibetan herbal preparation: Cordyceps sinensis. Int J Radiat Biol. 2008;84:139-149.

87.

Kim

Lee

Kim

. Protective effect of an herbal preparation (HemoHIM) on radiation-induced intestinal injury in mice. J Med Food. 2009;12:1353-1358.

88.

Han

Zhou

Zhang

. Angelica sinensis down-regulates hydroxyproline and Tgfb1 and provides protection in mice with radiation-induced pulmonary fibrosis. Radiat Res. 2006;165:546-552.

89.

Tian

Wang

. Rectal radiation injuries treated by Shen Ling Bai Zhu powders combined with rectal administration of western drugs [in Chinese]. Zhongguo Zhong Xi Yi Jie He Za Zhi. 2008;28:159-160.

90.

Dou

Yang

Wei

. The study of early application with Dixiong Decoction (地芎汤) for non-small cell lung cancer to decrease the incidence and severity of radiation pneumonitis: a prospective, randomized clinical trial. Chin J Integr Med. 2010;16:411-416.

91.

Kikawa

Herrick

Tateo

. Induced oxidative stress and cell death in the A549 lung adenocarcinoma cell line by ionizing radiation is enhanced by supplementation with docosahexaenoic acid. Nutr Cancer. 2010;62: 1017-1024.

92.

Lee

Kinniry

Arguiri

. Dietary curcumin increases antioxidant defenses in lung, ameliorates radiation-induced pulmonary fibrosis, and improves survival in mice. Radiat Res. 2010;173:590-601.

93.

Yildiz

Soyuer

Saraymen

. Protective effects of caffeic acid phenethyl ester on radiation induced lung injury in rats. Clin Invest Med. 2008;31:E242-E247.

94.

Bese

Munzuroglu

Uslu

. Vitamin E protects against the development of radiation-induced pulmonary fibrosis in rats. Clin Oncol (R Coll Radiol). 2007;19:260-264.

95.

Sener

Jahovic

Tosun

. Melatonin ameliorates ionizing radiation-induced oxidative organ damage in rats. Life Sci. 2003;74:563-572.

96.

Mantovani

Maccio

Melis

. Restoration of functional defects in peripheral blood mononuclear cells isolated from cancer patients by thiol antioxidants alpha-lipoic acid and N-acetyl cysteine. Int J Cancer. 2000;86:842-847.

97.

Beardsley

Pierschbacher

Wetzel

. Induction of T-cell maturation by a cloned line of thymic epithelium (TEPI). Proc Natl Acad Sci U S A. 1983;80: 6005-6009.

98.

Cerchietti

Navigante

Castro

. Effects of eicosapentaenoic and docosahexaenoic n-3 fatty acids from fish oil and preferential Cox-2 inhibition on systemic syndromes in patients with advanced lung cancer. Nutr Cancer. 2007;59:14-20.

99.

Lissoni

Chilelli

Villa

. Five years survival in metastatic non-small cell lung cancer patients treated with chemotherapy alone or chemotherapy and melatonin: a randomized trial. J Pineal Res. 2003;35:12-15.

100.

Norsa

Martino

. Somatostatin, retinoids, melatonin, vitamin D, bromocriptine, and cyclophosphamide in advanced non-small-cell lung cancer patients with low performance status. Cancer Biother Radiopharm. 2006;21:68-73.

101.

Lissoni

Tisi

Barni

. Biological and clinical results of a neuroimmunotherapy with interleukin-2 and the pineal hormone melatonin as a first line treatment in advanced non-small cell lung cancer. Br J Cancer. 1992;66: 155-158.

Lung Cancer Survival With Herbal Medicine and Vitamins in a Whole-Systems Approach

Abstract

Keywords

Background

Randomized Controlled Trials in CAM Therapies

Objectives

Methods

Patients

Human Subjects Concerns

Treatment Rationale

Diet

Exercise

Data Sources

Stage

Analysis Methods

Data Analysis

Manuscript Preparation

Results

Number of Patients

Comparing Long-term With Short-term PAM+V Use

Comparing Long-term PAM+V-Treated Patients With Concurrent External Controls From the California Cancer Registry

Comparing Long-term PAM+V-Treated Patients With Concurrent External Controls From Kaiser Permanente

Comparing All PAM+V Users With Concurrent External Controls From the California Cancer Registry

Comparing All PAM+V-Treated Patients With Concurrent External Controls From Kaiser Permanente

Comparing Combinations of PAM+V With Chemotherapy, Surgery, and Radiotherapy

Distinguishing the Effect of Treatment From Location of Follow-up

CAM Therapy in External Controls

Validity of External Controls

Crude Survival Rates: Stage II

Crude Survival Rates: Stage IIIA

Crude Survival Rates: Stage IIIB

Crude Survival Rates: Stage IV

Discussion

Principal Findings

Strengths of the Study

Limitations of the Study

Causal Inference Methods Used in Our Study

Possible Mechanisms

General cancer cell cycle effects

Chemotherapy-related effects

Reduction of chemotherapy toxicity and side effects

Reduction of radiation toxicity and side effects

More general systemic effects such as immune system modulation

Summary and Future Directions

Footnotes

Acknowledgements

References