Abstract

Patients exhibiting nasal irritation symptoms (NIS) during chemotherapy for breast cancer were recruited to identify the causative agents and risk factors for NIS. The regimens, anticancer agents used, doses administered, gender, age, body surface area, serum creatinine level (Cr), AST, ALT, performance status, clinical stage, and presence or absence of NIS were investigated using the medical charts of breast cancer patients who underwent chemotherapy between July 2008 and November 2010. To identify the causative agents of NIS, Fischer’s exact test was performed for each anticancer agent used in relation to the presence or absence of NIS. To select candidate risk factors for NIS, patients administered anticancer agents suspected of causing NIS were selected, classified into those with and without NIS, and univariate analysis was performed for each factor. Multiple logistic regression analysis was performed using factors with a P value of less than 0.1, dose of the causative agent, and AST, an index of liver function, as explanatory variables and the presence or absence of NIS as the objective variable, and the effects of candidate factors were evaluated. The cutoff values of risk factors were determined by receiver operating characteristic analysis. The anticancer agents administered to 77 target patients were cyclophosphamide (CPA), epirubicin, 5-FU, trastuzumab, paclitaxel, and docetaxel, and a significant difference was observed in the incidence of NIS in the CPA group only (P = 0.005). NIS was noted in 7 of the 38 patients in the CPA group. Multiple logistic regression analysis using age as a candidate risk factor, CPA dose, and AST showed a significant difference according to age alone (P = 0.03, odds ratio: 0.88, 95% confidence interval: 0.79-0.98), and its cutoff value was 50 years. CPA may be identified as a causative agent of NIS, and age (≤50 years) could be a risk factor. The anxiety of younger patients awaiting the administration of CPA may be alleviated by providing information that includes NIS. As this is a preliminary survey, it may be necessary to conduct a randomized, controlled comparative study, in order to identify agents and risk factors associated with NIS.
Key words Breast cancer, Chemotherapy, Nasal irritation symptoms, Cyclophosphamide, Adverse drug reaction

Author and Article Information

Author info
1. Department of Medical Oncology, Toyama University Hospital
2. Department of Pharmaceutical Therapy & Neuropharmacology, Faculty of Pharmaceutical Sciences, Graduate School of Medicine and Pharmaceutical Sciences, University of Toyama
3. Department of Pharmacy, Tonami General Hospital
4. Department of Pharmacy, Toyama Red Cross Hospital
5. Department of Pharmacy, Saiseikai Takaoka Hospital
6. Department of Surgery, Saiseikai Takaoka Hospital

RecievedJan 10 2014　　AcceptedApr 2 2014　PublishedApr 9 2014

CitationTanabe K, Takahata H, Takata K, Ikezaki T, Ohkubo J, Takeuchi M et al. (2014) Causative agents and risk factors for nasal irritation symptoms observed during chemotherapy in breast cancer patients. Science Postprint 1(1): e00017. doi: 10.14340/spp.2014.04C0001

Copyright©2013 The Authors. Science Postprint published by General Healthcare Inc. This is an open access article under the terms of the Creative Commons Attribution-NonCommercial-NoDerivs 2.1 Japan (CC BY-NC-ND 2.1 JP) License, which permits use and distribution in any medium, provided the original work is properly cited, the use is non-commercial and no modifications or adaptations are made.

FundingWe have received support from no external funding source for this study, because we have performed this study using an operating fund issued by our organization.
The funder had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

Competing interestsNone

Donation messageWe wish to contribute to advances in safe and reliable techniques of cancer chemotherapy. We are awaiting the generous support from those who have read this report.

Ethics statementIn ethical consideration of the patients evaluated, data processing was performed with unlinkable anonymization. As this study was an observation study not exceeding the range of daily clinical activities, informed consent was assumed to have been obtained as general consent by bulletin board posting in the hospital and explicit statements in the objectives of the use of clinical information. Therefore, this study was performed with the consent of the Institutional Review Board of Saiseikai Takaoka Hospital and Toyama University Hospital, Toyama Prefecture.

Corresponding authorKouichi Tanabe
AddressDepartment of Medical Oncology, Toyama University Hospital
2630 Sugitani, Toyama city, Toyama prefecture, 930-0194, Japan
E-mailIf you want to contact author,Please register as a member.

Introduction

Many regimens are currently regarded as standard chemotherapies for breast cancer 1, and various anticancer agents are used in these regimens 2. Therefore, adverse drug reactions and the measures taken to control them also vary 3-5. However, the possibility that some mild adverse drug reactions remain unreported cannot be excluded because the detection rate of these reactions may differ markedly depending on the investigation method used. For example, a previous study suggested that the detection rate of dysgeusia varied with the timing of the onset of symptoms and method of medical intervention used 6. Furthermore, although the incidence of malaise is known to be high 7, patients have infrequently reported this to their physicians 8. Outpatients in particular may return to their homes with lingering anxiety about even mild adverse drug reactions unless the physician provides information about their management and outcome.
Identifying the causative agent of an unreported adverse drug reaction is of importance when a combination of multiple anticancer agents are used because it leads to the prompt development of measures appropriate for the symptoms observed and causative agent. We established the Outpatient Chemotherapy Center at Saiseikai Takaoka Hospital in Toyama Prefecture, and soon encountered patients exhibiting nasal irritation symptoms (NIS) among those undergoing chemotherapy for breast cancer. However, we could not find information in the literature concerning these symptoms, and, consequently, could not provide an effective intervention or relevant information including the course of the symptoms, causative agent, risk factors, or methods for their management. Unnecessary anxiety may be avoided if these factors can be determined by selecting patients at high risk of NIS and providing an appropriate information service. Therefore, we performed the present study to identify the causative agents of and risk factors for NIS.

Methods

Overview

This study was designed as two-step research because NIS was an unknown adverse event. It consisted of 1) a prospective cohort study to identify the causative agent of NIS and 2) a retrospective cohort study performed in series to evaluate the risk factors for NIS.

Subjects

All subjects were patients who underwent chemotherapy for breast cancer at the Outpatient Chemotherapy Center at Saiseikai Takaoka Hospital in Toyama Prefecture between July 2008 and November 2010. The presence or absence of NIS, chemotherapy course in which the symptoms occurred, times until the appearance and disappearance of these symptoms from the beginning of the drip infusion of the anticancer agent, name of the regimen, names of the anticancer agents administered, their doses, gender, age, body surface area, performance status (PS), clinical stage, serum creatinine level (Cr), AST, and ALT of each patient were recorded.

Onset, characteristics, and outcome of NIS

Patients were first examined by a physician, and only those who were considered capable of tolerating chemotherapy visited the Outpatient Chemotherapy Center and underwent chemotherapy. The presence or absence of NIS was assessed when vital signs were measured by nurses at the Outpatient Chemotherapy Center (at the beginning, 5, 15, 30, 60 minutes after the beginning, every 60 minutes thereafter, and at the end of the administration of each anticancer agent). The comments of patients were recorded as they were described (in words used by the patients) in their medical charts, if NIS were observed. However, comments were only recorded after 15 minutes when the drip infusion of the anticancer drug was continued, except at the end of the administration. Patients were requested to remain at the Outpatient Chemotherapy Center for 60 minutes following the end of the drip infusion of the anticancer drug (including the time for posttreatments such as postmedication and hemostasis and time until checkout) and to report if their symptoms had disappeared. The time of the disappearance of symptoms was recorded. The attending physician examined patients again if their symptoms had not disappeared by their departure, and necessary treatments such as the administration of medication were performed according to the physician’s judgment. Patients were asked whether these treatments were effective during their next visits and the time at which symptoms disappeared, and their answers were recorded in their medical charts as the outcome.

Causative agents

A cross-tabulation table of each administered anticancer agent was prepared to identify the agents responsible for NIS, and the chi-square test was performed concerning the presence or absence of NIS. Anticancer drugs that showed a significant difference were judged to be causative agents.

Risk factors

1) Patients administered an anticancer drug judged to be a causative agent were selected from the cohort and divided into NIS and no-NIS groups to identify candidate risk factors for NIS.
2) Univariate analysis was performed using the t-test regarding the dose of the anticancer agent judged to have caused NIS and the patient’s age, body surface area, AST, ALT, and Cr and using the Mann-Whitney U-test concerning the PS and clinical stage.
3) The effects of the candidate factors were evaluated by multiple logistic regression analysis using the factors that showed P <0.1 on univariate analysis, the dose of the anticancer drug judged to be a cause of NIS, and results of laboratory test items that are indices of the function of the organ involved in the metabolism of the anticancer agent as explanatory variables and the presence or absence of NIS as the objective variable.
4) Receiver operating characteristic (ROC) analysis was performed concerning the variables that showed significant differences on multiple logistic regression analysis, and their cutoff values were determined.

Statistical analysis

Fisher’s exact test was performed if cells with an expected value of less than 5 were detected with the chi-square test. The level of significance was 0.05, and the Holm method (Bonferroni step-down method) was used to adjust the level of significance for multiple comparisons.
The level of significance for multiple logistic regression analysis was set at 0.05. Multicollinearity was confirmed (r <0.8) by preparing a correlation matrix of a combination of various explanatory factors, and the objective variable was confirmed not to be in a state of complete separation using scatter-plot matrices.
If the area under the ROC curve (AUC) was 0.5 or greater, the point at which the sum of the sensitivity and specificity was the maximum was selected as the cutoff value.
All these statistical procedures were performed using SPSS version 22 (IBM Japan Ltd., Tokyo).

Results

Patient backgrounds

Seventy-seven patients participated in this study (Table 1), and 7 developed NIS. Eight regimens and 6 anticancer agents were used (Table 2).

Table 1Patient backgrounds (N = 77)

Abbreviations: BSA, body surface area; SCr, serum creatinine; AST, aspartate aminotransferase; ALT, alanine transaminase; ECOG, Eastern Cooperative Oncology Group; PS, Performance Status.

Table 2Onset of NIS in each regimen (N = 77)

Abbreviations: DTX, docetaxel; CPA, cyclophosphamide; EPI, epirubicin; 5-FU, 5-fluorouracil; PTX, paclitaxel; HER, trastuzumab; q3w, every three weeks.
*: Dosage units: HER is in mg/kg; all other drugs are in mg/m2
**: HER dosage: initial dose 8 mg/kg, subsequent doses 6 mg/kg
***: HER dosage: initial dose 4 mg/kg, subsequent doses 2 mg/kg

Causative agents

Of the 6 anticancer drugs used, a significant difference was only observed in the cyclophosphamide (CPA) group (P = 0.005), in which the incidence of NIS was 18.4% (Table 3).

Table 3Results of the search for causative drugs (N = 77)

Abbreviations: DTX, docetaxel; CPA, cyclophosphamide; EPI, epirubicin; 5-FU, 5-fluorouracil; PTX, paclitaxel; HER, trastuzumab.
*: Significant difference (Fisher's exact test after the Holm (Bonferroni Step-down) correction)

Risk factors

Univariate analysis revealed that only age was extracted as a candidate risk factor of NIS (P <0.1), and the NIS group was significantly younger (P = 0.009) (Table 4).

Table 4Results of the univariate analysis for NIS

Abbreviations: BSA, body surface area; CPA, cyclophosphamide; SCr, serum creatinine; AST, aspartate aminotransferase; ALT, alanine transaminase; ECOG, Eastern Cooperative Oncology Group; PS, Performance Status.
*: Candidate risk factors (P <0.1, Student's t-test or Mann-Whitney's U test)

Multiple logistic regression analysis was performed using the dose of CPA and AST as covariates in addition to age. AST, the P value of which was lower than that of ALT, was used as an index of liver function because CPA is converted to an active form as it is metabolized in the liver. Therefore, only age showed a significant difference (P = 0.03, odds ratio: 0.88, 95% confidence interval: 0.79-0.98) (Table 5).
ROC analysis (P = 0.01, AUC: 0.81, 95% CI: 0.66-0.97) revealed that the cutoff value of age was 50 years.

Table 5Results of the multiple logistic regression analysis

Abbreviations: CPA, cyclophosphamide; AST, aspartate aminotransferase.
*Significant difference (P <0.05)

Patterns of the onset, characteristics, and outcome of NIS

NIS appeared in all patients near the end of the CPA administration and disappeared between approximately 1 hour after CPA was administered and within the day of its administration (Table 6).
A symptom described by all patients who developed NIS was “a stinging stimulus in the nostrils similar to that of wasabi (Japanese horseradish)”; however, the intensity expressions of the stimulation varied from “slight” to “strong enough to cause lacrimation”.
Regarding the treatment administered, 2 patients were suspected of being allergic to the anticancer drug by the attending physician and were administered diphenhydramine tablets (10 mg x 5 tablets) once. While marked drowsiness was noted in both patients, no change was observed in the time until the resolution of NIS or intensity of its symptoms, and the treatment was judged to be ineffective.

Table 6Onset of NIS and recovery patterns

Abbreviation: CPA, cyclophosphamide dosage
a: 4 courses in total, b: After initiating the administration of CPA, c: After the end of its administration, d: Treatment upon the onset of NIS and effects thereof, e: At the end of the administration of CPA (its administration time was 30 min)
The time of NIS disappearance was unknown in cases 1, 2, and 6 because the patients forgot to record the disappearance time after returning home.

Discussion

This is the first study to confirm NIS as an adverse event associated with chemotherapy for breast cancer and determine its causative agents and risk factors.
The most important finding of this study was the identification of CPA as the causative agent of NIS, which was derived from CPA being the only anticancer drug administered to all patients who developed NIS and also NIS occurring shortly after the beginning of its administration. However, a metabolite of CPA may be involved in the etiology of NIS because CPA is considered to be odorless 9. For example, acrolein, a metabolite of CPA, is known to be a cause of hemorrhagic cystitis, which has been associated with the administration of CPA 10, 11. However, previous studies demonstrated that acrolein was excreted in expired gas 12, possessed strong mucosal irritant properties 13, and was typically detoxified by glutathione, the rate-regulating enzyme of acrolein excretion, and caused tissue damage at exposed sites depleted of glutathione 13. These findings suggested that acrolein may be a cause of NIS.
The results obtained in the present study also revealed that age (≤50 years) is an independent risk factor for NIS. Although the risk of adverse events generally increases with age, the risk of NIS was higher in younger patients. The reason for this remains unclear, but may be explained by the following hypothesis: CYP2B6 is considered to be the rate-regulating enzyme in the conversion of CPA, a prodrug, into an active form 11, and its levels have been shown to decrease with aging 14. A previous study demonstrated that the clearance of propofol, which is primarily metabolized by CYP2B6, decreased with aging (≥65 years) 15. These findings suggest that NIS are less likely to occur in older individuals due to delays not only in the conversion of CPA into an active form by hepatic CYP2B6, but also the subsequent conversion of its final metabolite into acrolein.
NIS occurred in all scheduled courses of chemotherapies, rapidly disappeared in 5 of the 7 patients who developed then, and this pattern was reproducible. These results indicate that NIS is not an incidental phenomenon.
This study identified the causative drug and risk factors for NIS. On the other hand, it had limitations that need to be evaluated further: First, only a small number of breast cancer patients participated in the present study. Second, it was performed at a single facility. Third, there may be other risk factors that could not be identified by this study. Fourth, other risk factors may include, for example, a CYP2B6 gene polymorphism 16, 17 and smoking, which induces the expression of CYP2B6 18, because the amount and activity of CYP2B6 have been shown to affect the risk of mucosal damage such as hemorrhagic cystitis and oral mucositis 19. Fourth, the dose of CPA cannot be excluded from the risk factors for NIS because other anticancer regimens using different doses of CPA were not evaluated in this study. Finally, although other factors, such as previously administered concomitant drugs, media, and additives, may also be responsible for NIS, it was necessary to focus on a limited number of factors in consideration of the number of accumulated cases.
Details of NIS are unknown; therefore, it has been difficult to inform patients about it, in advance. This may have resulted in the occurrence of NIS in some cases as an unexpected adverse drug effect following the physician’s examination at the beginning of the administration of anticancer drugs and caused unnecessary anxiety to patients. Furthermore, physicians have not been able to administer evidence-based responsible treatments due to this lack of information. Therefore, despite the limitations described above, this study, in which details on the occurrence of NIS were clarified and the identification of high-risk patients was made possible, is considered to provide insight into solutions to these clinical problems. Moreover, research may develop further by investigating the relationships between NIS and similar, but more important mucosal disorders such as hemorrhagic cystitis and oral mucositis, and examining whether NIS serves as a predictive factor of these adverse events.
In conclusion, this study clarified that CPA may be a causative agent of NIS, and identified young age (≤50 years) as a risk factor. Therefore, providing information regarding NIS to younger patients prior to the administration of CPA, should mitigate their anxiety. As this is a preliminary survey, it may be necessary to conduct a randomized, controlled comparative study, in order to identify agents and risk factors associated with NIS.

Acknowledgments

Some of the results of this study were presented at the 10th International Conference of the Asian Clinical Oncology Society (ACOS 2012).

Author contributions

Tanabe K: Designed the study and wrote the manuscript.
Takahata H, Takata K: Cooperated to design the study and analyzed the data.
Ikezaki T, Ohkubo J, Takeuchi M, and Handa A: Interviewed patients, instructed the use of drug, and reviewed and edited the discussion section.
Hirutani K, Nitta A, Kashii T, and Murakami N: Advised on the study design and contributed to the introduction and discussion section by reviewing the manuscript.
Kitazawa H: Directed the entire process, from study design to writing the manuscript.

References

1. National Comprehensive Cancer Network (2013) NCCN Clinical Practice Guidelines in Oncology. Breast Cancer (version 3.2013).
2. Bergh J, Jonsson PE, Glimelius B, Nygren P, Brandt L, Brorsson B et al. for the SBU-group (2001) A systematic overview of chemotherapy effects in breast cancer. Acta Oncol 40: pp. 253-81.  doi: 10.1080/02841860120784.
3. Butters DJ, Ghersi D, Wilcken N, Kirk SJ, Mallon PT (2010) Addition of drugs to a chemotherapy regimen for metastatic breast cancer. Cochrane Database Syst Rev 11: pp. 1-72. doi: 10.1002/14651858.CD003368.pub3.
4. Cardoso F, Loibl S, Pagani O, Graziottin A, Panizza P, Martincich L et al. (2012) The European Society of Breast Cancer Specialists recommendations for the management of young women with breast cancer. Eur J Cancer 48: pp. 3355-77. doi: 10.1016/j.ejca.2012.10.004.
5. Carroll J, Protani M, Walpole E, Martin JH (2012) Effect of obesity on toxicity in women treated with adjuvant chemotherapy for early-stage breast cancer: a systematic review. Breast Cancer Res Treat 136: pp. 323-30. doi: 10.1007/s10549-012-2213-3.
6. Ishikawa T, Morita J, Kawachi K, Tagashira H (2013) Incidence of dysgeusia associated with chemotherapy for cancer [Article in Japanese]. Gan To Kagaku Ryoho 40: pp. 1049-54.
7. Tanabe K, Murasaki Y, Takaki A, Yoshida T, Nunotani M, Takahata H et al. (2009) Usefulness of Palliative Care Screening by Means of Seirei Mikatahara Hospital’s Support Team Assessment Schedule (STAS). Jpn J Pharm Health Care Sci 35: pp. 806-17. doi: 10.5649/jjphcs.35.806.
8. Passik SD, Kirsh KL, Donaghy K, Holtsclaw E, Theobald D, Cella D et al. (2002) Patient-related barriers to fatigue communication: initial validation of the fatigue management barriers questionnaire. J Pain Symptom Manage 24: pp. 481-93. doi: 10.1016/S0885-3924(02)00518-3.
9. Shionogi & Co. Ltd. (ed.) (2008) Description of active ingredient [in Japanese]. In: The drug interview form of Endoxan®. Shionogi & Co. Ltd., Osaka, pp 4-5.
10. Ramu K, Fraiser LH, Mamiya B, Ahmed T, Kehrer JP (1995) Acrolein mercapturates: synthesis, characterization, and assessment of their role in the bladder toxicity of cyclophosphamide. Chem Res Toxicol 8: pp. 515-524.
11. Brade WP, Herdrich K, Varini M (1985) Ifosfamide-pharmacology, safety and therapeutic potential. Cancer Treat Rev 12: pp. 1-47.
12. Parent RA, Caravello HE, Sharp DE (1996) Metabolism and distribution of [2,3-14C] acrolein in Sprague-Dawley rats. J Appl Toxicol 16: pp. 449-457. doi: 10.1002/(SICI)1099-1263(199609)16:5<449::AID-JAT369>3.0.CO;2-9.
13. Beauchamp RO Jr, Andjelkovich DA, Kligerman AD, MorganKT, Heck HD (1985) A critical review of the literature on acrolein toxicity. Crit Rev Toxicol 14: pp. 309-380. doi: 10.3109/10408448509037461.
14. Kudo T, Ito K (2013) Drug metabolizing enzymes and transporters  [Article in Japanese]. Respiration Res 32: pp. 439-446.
15. Kansaku F, Kumai T, Sasaki K, Yokozuka M, Shimizu M, Tateda T et al. (2011) Individual differences in pharmacokinetics and pharmacodynamics of anesthetic agent propofol with regard to CYP2B6 and UGT1A9 genotype and patient age. Drug Metab Pharmacokinet 26: pp. 532-537. doi: 10.2133/dmpk.DMPK-11-RG-039.
16. Manosuthi W, Sukasem C, Lueangniyomkul A, Mankatitham W, Thongyen S, Nilkamhang S et al. (2013) Impact of pharmacogenetic markers of CYP2B6, clinical factors, and drug-drug interaction on efavirenz concentrations in HIV/tuberculosis-coinfected patients. Antimicrob Agents Chemother 57: pp. 1019-1024. doi: 10.1128/AAC.02023-12.
17. Kharasch ED, Stubbert K (2013) Role of cytochrome P450 2B6 in methadone metabolism and clearance. J Clin Pharmacol 53: pp. 305-313. doi 10.1002/jcph.1.
18. Washio I, Maeda M, Sugiura C, Shiga R, Yoshida M, Nonen S et al. (2011) Cigarette smoke extract induces CYP2B6 through constitutive androstane receptor in hepatocytes. Drug Metab Dispos 39: pp. 1-3. doi: 10.1124/dmd.110.034504.
19. Rocha V, Porcher R, Fernandes JF, Filion A, Bittencourt H, Silva W Jr et al. (2009) Association of drug metabolism gene polymorphisms with toxicities, graft-versus-host disease and survival after HLA-identical sibling hematopoietic stem cell transplantation for patients with leukemia. Leukemia 23: pp. 545-556. doi: 10.1038/leu.2008.323.