Effects of Complementary and Alternative Medicine on Chemotherapy Delivery in Thai Patients

Background: Complementary and Alternative Medicine (CAM) is widely used among cancer patients worldwide. This prospective observational study aimed to show the effect of CAM use on chemotherapy delivery in Thai patients. Methods: During March 2014 to February 2015, the patients with breast, lung or colorectal cancer receiving first cycle chemotherapy at King Chulalongkorn Memorial Hospital were enrolled. The correlation between CAM using and chemotherapy schedule delay and dose reduction, dose intensity, quality of life and adverse event rates were analyzed. Results: There were 80 (44.20%) patients using CAM among 181 enrolled patients. Seventy six CAM users and 97 non-CAM users receiving 2nd cycle of chemotherapy were included for primary analysis. The chemotherapy schedules were delayed and/or reduced in 40 (52.6%) and 48 (49.5%) in CAM users and non-CAM users, respectively, p =0.681. The mean relative dose intensity (RDI) were 92.4% and 94.1% in CAM and non-CAM users, respectively, p=0.244. However, there were significantly more CAM users receiving chemotherapy less than 90% RDI (34.8% vs 19.8%, p=0.033). As compared to first cycle, at third cycle, the mean QOL score changes were -4.63 (95% CI -2.49-9.27) and -8.02 (-2.36- 9.142) in CAM user and non-CAM user, respectively (p=0.255). There were significantly higher rates of grade 3 or 4 anemia (5.1% vs 0%, p=0.024), and grade 2 malaise (19.0% vs 5.1%, p=0.004) in CAM users. Conclusions: There were similar overall rates of chemotherapy schedule delay and dose reduction between CAM- and non-CAM users. However, there were less CAM-users achieving 90% chemotherapy RDI.


Introduction
Complementary and Alternative Medicine (CAM) is a medical product or practice that is used together with or instead of standard medical care. It is widely used among cancer patients worldwide, ranging from 30-60% (Shih et al., 2009;Akyol and Oz, 2011;Naing et al., 2011;Horneber et al., 2012;Klafke et al., 2012;Puataweepong et al., 2012;Oyunchimeg et al., 2017;Jones et al., 2019;Sanford et al., 2019;Wode et al., 2019;Razali, 2020). In Thailand, one study reported CAM use in 60.9% of cancer patients. The major reasons for CAM use in cancer patients are to improve wellbeing and immunity, relief conventional treatment toxicity and treat cancer (Shih et al., 2009;Puataweepong et al., 2012;Smith et al., 2014;Oyunchimeg et al., 2017;Razali, 2020). However, most reports demonstrated patient satisfaction and benefit of CAM in term of quality of life and spiritual well-being improvement (Chandwani et al., 2010;Mao et al., 2010;(Zeller et al., 2013;Drozdoff et al., 2019). Although many patients expect to improve chemotherapy tolerability and treatment outcomes by using CAM, the potential harm of CAMs might preclude these expected benefit. Although, some studies showed protective effects of CAMs for chemotherapy induced leucopenia, they did not demonstrate how CAMs affect chemotherapy delivery (Zhuang et al., 2009;Zhuang et al., 2012;Jia et al., 2015). We conducted this prospective observational study to explore the effect of CAM on chemotherapy treatment delivery in Thai cancer patients receiving chemotherapy.

Patients
The patients with breast, lung or colorectal cancer receiving first cycle of standard dose adriamycin plus cyclophosphamide, carboplatin plus paclitaxel or gemcitabine and capecitabine plus oxaliplatin regimens, respectively, at the King Chulalongkorn Memorial Hospital (KCMH) were enrolled. All patients provided written informed consent before being enrolled to the study.
We conducted the face-to-face interviews for each patient to collect the CAM use data. The interviewers were the trained health care personnel not-involving in patient cancer treatments. After an eligible patient providing the informed consent, two interviews were done on the day patient receiving first and third cycles of chemotherapy. The data was reported only to investigator after completion of all treatment sessions in all patients. We assessed the quality of life by using FACT-G (Functional Assessment of Cancer Therapy -General) permitted by FACIT.org. The chemotherapy schedule and dose modifications, and adverse events were collected from medical records by investigators. The chemotherapy schedule delays and dose reductions were assessed in second and third chemotherapy cycles . Chemotherapy delay was defined of any postpone of chemotherapy planned schedule and dose reduction was defined as any dose reduction in subsequent cycle from actual dose in first cycle. We calculated chemotherapy dose intensity in all patients receiving 4 cycles of chemotherapy. Dose intensity is unit dose of chemotherapy administered per unit time. Relative dose intensity (RDI) was defined as actual dose divided by the standard dose.
The definition of Complementary and Alternative Medicine (CAM) in this study was any medical product that patient put into body per oral, per intravenous injection or per rectal on purpose for treatment, that is not thought of as standard care. The severity of adverse events was assessed based on the Common Terminology Criteria for Adverse Events, version 3.0.

Statistical Analysis
The co-primary end points were the rates of schedule delay and/or dose reduction, and dose intensity of chemotherapy. All primary endpoints were assessed based on chemotherapy delivery in first three cycles. According to our chemotherapy administration database, the rates of chemotherapy schedule delay and/or dose reduction was around 30%. The 25% of difference in chemotherapy delay and/or reduction rates between CAM and non-CAM users were considered clinical significance. With 80% power to detect that difference, we required 93 patients for each group. We compared the rates of chemotherapy schedule delay and dose reduction, rates of RDI less than 90% and adverse event rates between CAM users and non-CAM users by Chi-squared test. 95% confidence intervals for the chemotherapy delivery outcomes were calculated by bootstrap method. We used t-test to compare chemotherapy dose intensity between CAM users and non-CAM users. P values <0.05 was considered statistically significant. All statistical analyses were performed using SPSS version 16.0 for Windows software (SPSS Inc., Chicago, IL). cyclophosphamide. Two third of patients had localized or loco-regional disease. There were 80 (44.2%) and 101 (55.8%) Cam users and non-CAM users, respectively. More patients lived in capital city, Bangkok, in CAM users compared to non-CAM users (57.5% vs. 34.7%, p=0.019).

Between
Although not statistically significant, there were more patients with breast cancer in non-CAM users. Baseline characteristics in CAM users and non-CAM users were shown in Table 1.
To improve efficacy, decrease toxicity, and decrease worry and fear of conventional therapy were the most common reason for CAM use in this study. Herbal medicine is the most common type of CAM used among our patients. Almost all patients did not complaint any adverse event related to CAM use. Most patients (71.3%) did not inform their doctors regarding the CAM use. The CAM use detail was shown in Table 2.
Among CAM users, there were 76 (95%) and 72 (90%) patients receiving second and third cycles, respectively. There were 97 (96%) and 92 (91%) non-CAM users receiving second and third cycles, respectively.The reasons for not receiving the subsequent cycles were death and/or disease progression in five and two patients in CAM users and non-CAM users, respectively. Three CAM users and seven non-CAM users did not show up on the subsequent chemotherapy dates. Among 173 patients receiving at least two cycles of chemotherapy, the chemotherapy was delayed and/or reduced in 40 (52.6%) CAM users and 48 (49.5%) non-CAM users, p =0.681. However, there were the trends toward more CAM users having schedule delay for more than 2 times, schedule delay longer than 14 days and more than 20% dose reduction. The chemotherapy schedule delay and     Table 3. Among 69 (86%) and 91 (90%) CAM users and non-CAM users receiving four cycles of chemotherapy, the mean RDI was 92.3% and 94.1% in CAM and non-CAM users, respectively, p=0.244. However, as compared with non-CAM users, there were significantly more CAM users receiving chemotherapy less than 90% RDI (34.8% vs 19.8%, p=0.033). The chemotherapy RDI was shown in Table 4.
For quality of life assessment, there were 43 and 48 patients in CAM and non-CAM users interviewed at first and third cycles. As compared to first cycle, at third cycle, the mean QOL score was -4.63 (95% CI -2.49-9.27) and -8.02 (-2.36-9.142) in CAM user and non-CAM user, respectively (p=0.255). There were 28 (65.1%) CAM users and 36 (75.0%) non-CAM users having decreased quality of life score at third cycle (p=0.303).
Overall adverse event rates in both groups were similar. As compared to non-CAM users, more CAM users developed grade 3 or 4 anemia (5.1% vs 0, p=0.024). There was a trend toward more increased aminotransferase in CAM users compared to non-CAM users. CAM users had significantly more grade 2 malaise than non-CAM users, (19.0% vs 5.1%, p=0.004). The adverse event data was shown in Table 5.

Discussion
In this prospective study, we compared chemotherapy treatment delivery between CAM users and non-CAM users. There was no difference in overall rates of chemotherapy schedule delay or dose reduction between these two groups. However, there were significantly more patients receiving chemotherapy with less than 90% relative doe intensity in CAM users. As compared to non-CAM users, CAM users had higher rates of some adverse events during chemotherapy.
Among 181 cancer patients receiving chemotherapy, 44.2% of patients were using CAM. It was less than 60.9% of Thai cancer patients receiving radiotherapy in previous report. (Puataweepong et al., 2012) Besides the population difference, the face-to-face interviewing in the chemotherapy center in this study might result in the less number of patients reporting CAM use than using self-report method in previous study.
The overall rates of chemotherapy schedule delay and dose reduction, and mean relative dose intensity were not different between CAM and non-CAM users. However, there were significantly more patients receiving chemotherapy less than 90% RDI in CAM users. This was likely related to a trend toward more major chemotherapy modification including more-than-2times or longer-than-14-day schedule delay or more than  Therefore, interpretation of these post-hoc analyses should be cautious with this potential confounding factor. To our knowledge, this is the first study comparing the chemotherapy treatment deliveries between CAM and non-CAM users.
As compared to non-CAM users, there were more incidences in certain adverse events including transaminitis, anemia and malaise in CAM users. In CAM users, the incidences of increased aminotransferases were 24-30% compared to 12% in non-CAM users, this might be related to hepatotoxicity of herbal medicine as shown in several reports (Teo et al., 2016;Philips et al., 2019). However, severe transaminitis was quite rare in CAM users in this study. For myelosuppreaaion, there were significantly more severe anemia in CAM users but no difference in neutropenia or thrombocytopenia. Differently, CAM protective effects on chemotherapy induced neutropenia was demonstrated in previous randomized trials and a systemic review (Zhuang et al., 2009;Zhuang et al., 2012;Jia et al., 2015). However, in this study, there was an imbalance in chemotherapy regimens and CAM types were unrestricted. Therefore, these findings should be cautiously interpreted.
In this study, around two third of CAM users did not disclose to their doctors. The CAM use disclosure rates were varied among previous reports. (Shih et al., 2009;Puataweepong et al., 2012;Wode et al., 2019) The major reason for that was no asking by their doctors. Given, the potential adverse effects of CAM in patients receiving chemotherapy, acquiring for CAM using information should be encouraged among oncologists.
Our study had some limitations including no chemotherapy schedule and dose modification protocol, various chemotherapy regimens and possible imbalance of patient characteristics including performance status and comorbidity. Also CAM users might not disclose their CAM use especially with the face-to-face interview method in the chemotherapy center. There was certainly a wide variation in CAM types used by the patients in this study. The study was a single center study, the generalizability might be limited.
In conclusion, this prospective study in Thai patients with cancer demonstrated similar overall rates of chemotherapy schedule delay and dose reduction between CAM users and non-CAM users receiving chemotherapy. However, there were significantly more CAM users receiving chemotherapy less than 90% relative dose intensity. Acquiring of CAM using data should be encouraged among the oncologists.