2016 Volume 39 Issue 11 Pages 1852-1858
There are individual differences in the frequency of chemotherapy-induced nausea and vomiting (CINV) in cancer patients. We investigated the individual variability in susceptibility to CINV with focus on both behavioral factors and genetic factors in Japanese cancer patients. We performed a prospective study to investigate the association between patient attributes (backgrounds and habits as well as gene polymorphisms) and anorexia, nausea, or vomiting in 55 Japanese cancer patients undergoing chemotherapy at Nagoya University Hospital. We found that gender (female), use of non-steroidal anti-inflammatory drugs, susceptibility to motion sickness, and anxiety were associated with the frequency of CINV. Gene polymorphisms of rs1076560 (dopamine D2 receptor gene), rs6766410 (serotonin 5-HT3C receptor gene) and rs4680 (catechol-O-methyltransferase gene) were also associated. Our data suggest that these attributes may thus be risk factors for CINV. Our results provide novel information that can be used to predict the incidence of CINV in Japanese patients undergoing chemotherapy; this can help provide a substantial improvement in supportive care for patients with different types of cancer.
Chemotherapy-induced nausea and vomiting (CINV) is one of the most distressing side effects of chemotherapy. Inadequately controlled emesis impairs functional activity and QOL for patients, increases the use of health care resources, and may occasionally compromise adherence to treatment.1,2) According to the American Society of Clinical Oncology Clinical Practice Guidelines regarding antiemetic agents in oncology, a three-drug combination of a serotonin 5-hydroxytryptamine type 3 (5-HT3) receptor antagonist, dexamethasone, and aprepitant is recommended before chemotherapy treatments that have high emetic risks. If this is insufficient, substituting a high-dose of intravenous metoclopramide for the serotonin 5-HT3 receptor antagonist or adding a dopamine D2 receptor antagonist to the regimen is considered.3,4) In recent years, a new class of antiemetic therapies represented by neurokinin-1 (NK-1) receptor antagonists has been introduced into clinical practice. An NK-1 receptor antagonist, aprepitant, combined with dexamethasone and the serotonin 5-HT3 receptor antagonist ondansetron, have been shown to ameliorate CINV in both high and moderate emetic risk chemotherapy patients.4,5)
In spite of antiemetic therapy being administered according to the guidelines for CINV treatment, an incidence rate of 25–38% for delayed emesis and 55–60% for delayed nausea has been observed.4) The mechanisms underlying interpatient variability in response to treatment are not clear but are believed to be due, in part, to genetic factors. To that end, some polymorphisms in the serotonin 5-HT3 receptor genes were investigated for their relationship with CINV.6–8) There is also the possibility that polymorphisms of genes that play functional molecular or physiological roles, such as the NK-1 receptor, dopamine D2 receptor, and catechol-O-methyltransferase (COMT), are related to CINV.9–12) The likelihood that nausea and vomiting will develop after chemotherapy depends on other many factors as well. Behavioral (patient-related) risk factors, including female gender, young age, a history of alcohol intake, prior episodes of morning or motion sickness, impaired QOL, and previous experience of CINV are known to increase the risk for CINV.1,2) Additionally, treatments that cause gastrointestinal problems, such as radiotherapy, opioid and non-steroidal anti-inflammatory drugs (NSAIDs), have the potential to exacerbate nausea and vomiting.13–15) However, there has been little examination of the associations between behavioral factors or genetic factors and CINV in Japanese cancer patients.
In the present study, we examined the individual variability in the expression of CINV to identify behavioral factors (gender, age, drug dependency, radiation therapy, history of alcohol intake, susceptibility to motion sickness, anxiety) and genetic risk factors (polymorphisms in genes including serotonin 5-HT3A receptor, serotonin 5-HT3B receptor, serotonin 5-HT3C receptor, dopamine D2 receptor, NK-1 receptor, and COMT) related to the incidence of CINV in Japanese subjects. Revealing an association between these factors and the incidence or severity of CINV can provide a substantial improvement in supportive care for cancer patients by predicting the probability of CINV beforehand.
Our prospective observational study was performed at Nagoya University Hospital. We enrolled 55 Japanese cancer patients who received initial (or first course post-relapse) chemotherapy containing either cisplatin or carboplatin at the gynecology and respiratory medicine ward. The patients’ characteristics are shown in Table 1. The chemotherapy regimen included antiemetic agents (a serotonin 5-HT3 receptor antagonist and dexamethasone) according to Nagoya University Hospital procedures. Data from all patients (33 males, 22 females) were analyzed for the association between CINV on one hand, and gene polymorphisms and patient backgrounds on the other. However, only respiratory medicine patients (including 2 females, all 33 males) were investigated for the association between CINV and a history of alcohol intake, kinetosis, and anxiety.
| Number | 55 | |
| Gender | Male | 33 (60%) |
| Female | 22 (40%) | |
| Average age | 61.8 (±8.6) years old | |
| Clinical examination | ||
| eGFR | 82.7 (±18.9) mL/min/1.73 mm2 | |
| γGTP | 30.7 (±25.5) IU/L | |
| ALB | 3.5 (±0.4) g/dL | |
| Ca | 4.6 (±0.2) mEq/L | |
| HGB | 12.6 (±1.7) g/dL | |
| Neutrophils | 4.5 (±2.6)×103/µL | |
| CRP | 1.2 (±2.1) mg/dL | |
| Type of cancer | ||
| Ovarian cancer | 9 (16.4%) | |
| Cervical cancer | 2 (3.6%) | |
| Endometrial cancer | 9 (16.4%) | |
| Small-cell lung cancer | 5 (9.1%) | |
| Non-small-cell lung cancer | 24 (43.6%) | |
| Malignant pleural mesothelioma | 6 (10.9%) | |
| Regimen of cancer chemotherapy | ||
| CBDCA+PTX | 17 (30.9%) | |
| CBDCA+DTX | 2 (3.6%) | |
| CDDP+5-FU | 1 (1.9%) | |
| CDDP+PEM | 8 (14.6%) | |
| CDDP+GEM | 3 (5.4%) | |
| CDDP+VP-16 | 3 (5.4%) | |
| CDDP+VNR | 8 (14.6%) | |
| CDDP+TS-1 | 1 (1.9%) | |
| CBDCA+PEM | 3 (5.4%) | |
| CBDCA+GEM | 1 (1.9%) | |
| CBDCA+VP-16 | 2 (3.6%) | |
| CBDCA+PTX | 3 (5.4%) | |
| CBDCA+PTX+BEV | 3 (5.4%) | |
Values in average age and clinical examination indicated the mean±S.D. eGFR, estimated glomerular filtration rate; γGTP, γ-glutamyltransferase; ALB, albumin; Ca, calcium; HGB, hemoglobin; CRP, C-reactive protein; CBDCA, carboplatin; PTX, paclitaxel; DTX, docetaxel; CDDP, cisplatin; 5-FU, 5-fluorouracil; PEM, pemetrexed; GEM, gemcitabine; VP-16, etoposide; VNR, vinorelbine; TS-1, tegafur, gimeracil, and oteracil; BEV, bevacizumab.
The major exclusion criteria for this study were opioid use, vomiting within 24 h or antiemetic use within 48 h before chemotherapy, and the presence of health issue that could cause nausea or vomiting.16)
OutcomesAssociations of multiple factors with CINV were determined based on outcomes. Gender, age, drug dependency, and radiation therapy were determined from electronic medical records; history of alcohol intake, susceptibility to motion sickness, and anxiety were determined through interviews conducted by pharmacists. We evaluated anorexia in addition to nausea and vomiting as symptoms induced by chemotherapy according to a previous study.17) Anorexia, nausea, and vomiting as indices of CINV were graded by the National Cancer Institute Common Terminology Criteria for Adverse Events (CTCAE) version 4.0 on the worst day during the 10 d after commencing chemotherapy. We considered grades ≥2 for anorexia (oral intake altered without significant weight loss or malnutrition; oral nutritional supplements indicated) and nausea (oral intake decreased without significant weight loss, dehydration or malnutrition), and ≥1 for vomiting [1–2 episodes (separated by 5 min) in 24 h], as CINV.18)
Target Gene PolymorphismsWe analyzed 10 polymorphisms (Table 2). For the serotonin 5-HT receptor genes, we analyzed the serotonin 5-HT3A receptor gene (HTR3A: GenBank accession no. DQ050460), rs1062613 on the 5ʹ end untranslated region; serotonin 5-HT3B receptor gene (HTR3B: DQ050462), rs1176744 (p.Tyr129Ser) and rs145224778 (p.Ala223Thr) on the exon, and −100_−102 AAG deletion variant on the promoter region; and serotonin 5-HT3C receptor gene (HTR3C: AF459285), rs6766410 on the exon. The HTR3A and HTR3B genes are located on chromosome 11q23.1, while the HTR3C is located on chromosome 3q27. For the NK-1 receptor gene (TACR1: AY420417), we analyzed Y192H and rs17838409 on the exon. The TACR1 is located on chromosome 2p13.1–p12. For the dopamine D2 receptor gene (DRD2: AY418851), we analyzed rs6277 on the exon and rs1076560 on the intron. The DRD2 is located on chromosome 11q22–23. For the COMT gene (COMT: DQ040245), we analyzed rs4680 on the exon. The COMT is located on 22q11.21–23.
| Gene | Localization | SNP | Exchange | Gene mutation (%) | MAF | HWE p-value | ||
|---|---|---|---|---|---|---|---|---|
| −/− | +/− | +/+ | ||||||
| HTR3A | 5′ UTR | rs1062613 | C>T | 39 (71) | 14 (26) | 2 (4) | 0.16 | 0.85 |
| HTR3B | Promotor | −100_−102 AAG insertion/deletion | delAAG | 42 (76) | 12 (22) | 1 (2) | 0.13 | — |
| Exon 5 | rs1176744 | T>G | 22 (40) | 26 (47) | 7 (13) | 0.36 | 1.00 | |
| Exon 6 | Ala223Thr | G>A | 55 (100) | 0 (0) | 0 (0) | 0 | — | |
| HTR3C | Exon 5 | rs6766410 | A>C | 17 (30) | 27 (49) | 11 (20) | 0.44 | 1.00 |
| TACR1 | Exon 2 | Y192H | T>C | 0 (0) | 55 (100) | 0 (0) | 0 | — |
| Exon 4 | rs17838409 | G>A | 55 (100) | 0 (0) | 0 (0) | 0 | — | |
| DRD2 | Exon 7 | rs6277 | C>T | 43 (78) | 11 (20) | 1 (2) | 0.12 | 1.00 |
| Intron 6 | rs1076560 | C>A | 24 (44) | 26 (47) | 5 (9) | 0.33 | 0.87 | |
| COMT | Exon 4 | rs4680 | G>A | 31 (56) | 21 (38) | 3 (6) | 0.25 | 1.00 |
MAF, minor allele frequency; HWE, Hardy–Weinberg equilibrium; HTR3A, serotonin 5-HT3A receptor; HTR3B, serotonin 5-HT3B receptor; HTR3C, serotonin 5-HT3C receptor; TACR1, tachykinin receptor 1; DRD2, dopamine D2 receptor; COMT, catechol-O-methyltransferase.
DNA was collected from all patients. Single nucleotide polymorphisms were genotyped by the TaqMan® 5′-exonuclease allelic discrimination assay (Applied Biosystems, Foster City, CA, U.S.A.). We dispensed 1 µL of DNA samples (100 ng) in a 96-well reaction plate and added 0.25 µL of TaqMan®40X probe, 5 µL of TaqMan® Universal PCR Master Mix, and 3.75 µL of distilled water. DNA amplification and the measurement of fluorescence intensity of the target sequence-specific fluorogenic probes were performed by an Applied Biosystems 7300 Real Time PCR System. PCR conditions were as follows: after an initial denaturation at 95°C for 10 min, we programmed 50 cycles of 92°C for 15 s and 58°C for 1 min.
SequencingAll sequencing analysis was performed with PCR from genomic DNA amplification reactions. Reactions were performed in a total volume of 10 µL containing 100 ng of DNA, 2.5 mM deoxynucleoside triphosphate, 20 µM of each primer (Forward: 5′-GGG GCT CCT CTT AGA TTA CAT TAT TC-3′, Reverse: 5′-CAG ACG ATG AAT AAA AAC AAA C-3′) (RIKAKEN, Nagoya, Japan), 10× buffer, and rTaq DNA Polymerase using a PCR Thermal Cycler Dice® Standard (TaKaRa, Shiga, Japan). The initial denaturation was at 94°C for 1 min, followed by 40 cycles consisting of 94°C for 30 s, 55°C for 30 s, and 74°C for 1.5 min. A final extension period at 72°C for 7 min was performed. The PCR cycle sequencing program using a PCR thermal cycler consisted of an initial denaturation at 96°C for 1 min, followed by 25 cycles consisting of 96°C for 10 s, 50°C for 5 s, and 60°C for 4 min; the final extension period was at 72°C for 7 min. Analysis was performed using a Big-Dye® Primer cycle sequencing kit (Applied Biosystems) on an ABI PRISM® 310 Genetic Analyzer (Applied Biosystems).
Statistical AnalysesThe genotype and allele frequencies of each single nucleotide polymorphism, and the presence of Hardy–Weinberg equilibrium between expected and observed genotype distributions, were calculated using Haploview 4.2. The odds ratios (ORs) and 95% confidence intervals (CIs) were estimated using logistic regression using SPSS 23. p values less than 0.05 were considered statistically significant differences.
Ethical ConsiderationsThis study was approved by the ethics committee of Nagoya University Graduate School of Medicine and performed according to Good Clinical Practice Guidelines. The written informed consent documents were obtained from all subjects.
The frequency of anorexia according to each risk factor is shown in Table 3. The ORs for female gender and use of NSAIDs were significantly higher than those for male gender (OR, 9.6; 95% CI, 2.4–39; p<0.01) and no use of NSAIDs (OR, 6.0; 95% CI, 1.9–19; p<0.01). The OR of anxiety also indicated a higher tendency than that of not being anxiety (OR, 3.9; 95% CI, 0.83–18; p=0.085).
| Variable | Yes | No | p-Valuea) | Odds ratio | 95% Confidence interval | p-Valueb) |
|---|---|---|---|---|---|---|
| Gender (female) | 17/20 (85%) | 13/35 (37%) | 0.0010** | 9.6 | 2.4–39 | 0.0020** |
| Age (less than 50 years) | 5/6 (83%) | 25/49 (51%) | 0.13 | 4.8 | 0.52–44 | 0.17 |
| Radiation therapy (+) | 7/14 (50%) | 23/41 (56%) | 0.69 | 0.78 | 0.23–2.6 | 0.69 |
| Use of NSAIDs (+) | 7/28 (25%) | 9/27 (33%) | 0.0020** | 6.0 | 1.9–19 | 0.0030** |
| Use of aprepitant (−) | 21/39 (54%) | 9/16 (56%) | 0.87 | 1.1 | 0.34–3.6 | 0.87 |
| History of alcohol intake (−) | 13/22 (59%) | 4/13 (31%) | 0.55 | 0.64 | 0.15–2.7 | 0.55 |
| Prone to motion sickness (+) | 4/4 (100%) | 9/31 (29%) | 0.0060** | 3.9×109 | — | 1.0 |
| Anxiety state (+) | 6/10 (60%) | 7/25 (28%) | 0.077# | 3.9 | 0.83–18 | 0.085# |
| HTR3A: rs1062613 C/C | 23/39 (59%) | 7/16 (44%) | 0.30 | 1.8 | 0.57–6.0 | 0.31 |
| HTR3A: rs1062613 T/T | 2/2 (100%) | 28/53 (53%) | 0.19 | 1.4×109 | — | 1.0 |
| HTR3B: AAG deletion (−) | 22/42 (52%) | 8/13 (62%) | 0.56 | 1.5 | 0.40–5.2 | 0.56 |
| HTR3B: rs1176744 T/T | 14/22 (64%) | 16/33 (48%) | 0.27 | 1.9 | 0.62–5.6 | 0.27 |
| HTR3B: rs1176744 G/G | 4/7 (57%) | 26/48 (54%) | 0.88 | 1.1 | 0.23–5.6 | 0.88 |
| HTR3C: rs6766410 A/A | 11/17 (65%) | 19/38 (50%) | 0.31 | 1.8 | 0.56–6.0 | 0.31 |
| HTR3C: rs6766410 C/C | 7/11 (64%) | 23/44 (52%) | 0.50 | 1.6 | 0.41–6.2 | 0.50 |
| DRD2: rs6277 C/C | 22/43 (51%) | 8/12 (67%) | 0.34 | 0.52 | 0.14–2.0 | 0.35 |
| DRD2: rs6277 T/T | 1/1 (100%) | 29/54 (54%) | 0.36 | 1.4×109 | — | 1.0 |
| DRD2: rs1076560 C/C | 15/24 (63%) | 15/31 (48%) | 0.30 | 1.8 | 0.60–5.3 | 0.30 |
| DRD2: rs1076560 A/A | 3/5 (60%) | 27/50 (54%) | 0.80 | 1.3 | 0.20–8.3 | 0.80 |
| COMT: rs4680 G/G | 19/31 (61%) | 11/24 (46%) | 0.25 | 1.9 | 0.64–5.5 | 0.26 |
| COMT: rs4680 A/A | 0/3 (0%) | 30/52 (58%) | 0.051# | — | — | 1.0 |
“Yes” is percentage of patients expressed anorexia (≧ grade 2) on the worst day during 10 d after chemotherapy in patients with each risk factor. “No” is percentage of patients expressed anorexia (≧ grade 2) on the worst day during 10 d after chemotherapy in patients without each risk factor. ** p<0.01, # p<0.1 vs. corresponding “No” [a) chi-square test, b) univariate logistic regression analysis]. NSAID, non-steroidal anti-inflammatory drug; HTR3A, serotonin 5-HT3A receptor; HTR3B, serotonin 5-HT3B receptor; HTR3C, serotonin 5-HT3C receptor; DRD2, dopamine D2 receptor; COMT, catechol-O-methyltransferase.
The frequency of nausea according to each risk factor is shown in Table 4. The OR of being prone to motion sickness indicated a higher tendency than that of not being prone to motion sickness (OR, 10; 95% CI, 0.92–115; p=0.060). The OR of C/C homozygotes of rs1076560 (DRD2) was significantly higher than that of the A allele carriers (OR, 2.8; 95% CI, 0.92–8.4; p<0.05).
| Variable | Yes | No | p-Valuea) | Odds ratio | 95% Confidence interval | p-Valueb) |
|---|---|---|---|---|---|---|
| Gender (female) | 6/20 (30%) | 10/35 (29%) | 0.91 | 1.1 | 0.32–3.6 | 0.91 |
| Age (less than 50 years) | 1/6 (17%) | 15/49 (31%) | 0.48 | 0.45 | 0.49–4.2 | 0.49 |
| Radiation therapy (+) | 5/14 (36%) | 11/41 (27%) | 0.53 | 1.5 | 0.42–5.5 | 0.53 |
| Use of NSAIDs (+) | 9/28 (32%) | 7/27 (26%) | 0.61 | 1.4 | 0.42–4.4 | 0.61 |
| Use of aprepitant (−) | 9/39 (23%) | 7/16 (44%) | 0.13 | 2.6 | 0.75–8.9 | 0.13 |
| History of alcohol intake (−) | 6/22 (27%) | 4/13 (31%) | 0.83 | 1.2 | 0.26–5.3 | 0.83 |
| Prone to motion sickness (+) | 3/4 (75%) | 7/31 (23%) | 0.029* | 10 | 0.92–115 | 0.060# |
| Anxiety state (+) | 4/10 (40%) | 6/25 (24%) | 0.34 | 2.1 | 0.44–10 | 0.35 |
| HTR3A: rs1062613 C/C | 11/39 (28%) | 5/16 (31%) | 0.82 | 0.86 | 0.24–3.1 | 0.82 |
| HTR3A: rs1062613 T/T | 1/2 (50%) | 15/53 (28%) | 0.51 | 2.5 | 0.15–43 | 0.52 |
| HTR3B: AAG deletion (−) | 12/42 (29%) | 4/13 (31%) | 0.88 | 1.1 | 0.29–4.3 | 0.88 |
| HTR3B: rs1176744 T/T | 6/22 (27%) | 10/33 (30%) | 0.81 | 0.86 | 0.26–2.9 | 0.81 |
| HTR3B: rs1176744 G/G | 2/7 (29%) | 14/48 (29%) | 0.97 | 0.97 | 0.17–5.6 | 0.97 |
| HTR3C: rs6766410 A/A | 7/17 (41%) | 9/38 (24%) | 0.19 | 2.3 | 0.67–7.7 | 0.19 |
| HTR3C: rs6766410 C/C | 4/11 (36%) | 12/44 (27%) | 0.55 | 1.5 | 0.38–6.2 | 0.55 |
| DRD2: rs6277 C/C | 13/43 (30%) | 3/12 (25%) | 0.72 | 1.3 | 0.30–5.6 | 0.73 |
| DRD2: rs6277 T/T | 1/1 (100%) | 15/54 (28%) | 0.12 | 4.2×109 | — | 1.0 |
| DRD2: rs1076560 C/C | 11/24 (46%) | 5/31 (16%) | 0.016* | 4.4 | 1.3–15 | 0.020* |
| DRD2: rs1076560 A/A | 0/5 (0%) | 16/50 (32%) | 0.13 | — | — | 1.0 |
| COMT: rs4680 G/G | 10/31 (32%) | 6/24 (25%) | 0.56 | 1.4 | 0.43–4.7 | 0.56 |
| COMT: rs4680 A/A | 0/3 (0%) | 16/52 (31%) | 0.25 | — | — | 1.0 |
“Yes” is percentage of patients expressed nausea (≧ grade 2) on the worst day during the 10 d-after chemotherapy in patients with each risk factor. “No” is percentage of patients expressed nausea (≧ grade 2) on the worst day during 10 d after chemotherapy in patients without each risk factor. * p<0.05, # p<0.1 vs. corresponding “No” [a) chi-square test, b) univariate logistic regression analysis]. NSAID, non-steroidal anti-inflammatory drug; HTR3A, serotonin 5-HT3A receptor; HTR3B, serotonin 5-HT3B receptor; HTR3C, serotonin 5-HT3C receptor; DRD2, dopamine D2 receptor; COMT, catechol-O-methyltransferase.
The frequency of vomiting according to each risk factor is shown in Table 5. The ORs of being prone to motion sickness and C/C homozygosity of rs1076560 (DRD2) were significantly higher than those of not being prone to motion sickness (OR, 20; 95% CI, 1.6–245; p<0.05) and the A allele carriers (OR, 4.7; 95% CI, 1.1–20; p<0.05). The OR of anxiety also indicated a higher tendency than that of no anxiety (OR, 4.9; 95% CI, 0.85–28; p=0.075).
| Variable | Yes | No | p-Valuea) | Odds ratio | 95% Confidence interval | p-Valueb) |
|---|---|---|---|---|---|---|
| Gender (female) | 4/20 (20%) | 7/35 (20%) | 1.0 | 1.0 | 0.25–3.9 | 1.0 |
| Age (less than 50 years) | 0/6 (0%) | 11/49 (22%) | 0.19 | — | — | 1.0 |
| Radiation therapy (+) | 4/14 (29%) | 7/41 (17%) | 0.35 | 1.9 | 0.47–8.0 | 0.36 |
| Use of NSAIDs (+) | 6/28 (21%) | 5/27 (19%) | 0.79 | 1.2 | 0.32–4.5 | 0.79 |
| Use of aprepitant (−) | 7/39 (18%) | 4/16 (25%) | 0.55 | 1.5 | 0.38–6.2 | 0.55 |
| History of alcohol intake (−) | 5/22 (23%) | 2/13 (15%) | 0.60 | 0.62 | 0.10–3.8 | 0.60 |
| Prone to motion sickness (+) | 3/4 (75%) | 4/31 (13%) | 0.0030** | 20 | 1.6–245 | 0.018* |
| Anxiety state (+) | 4/10 (40%) | 3/25 (12%) | 0.061# | 4.9 | 0.85–28 | 0.075# |
| HTR3A: rs1062613 C/C | 8/39 (21%) | 3/16 (19%) | 0.88 | 1.1 | 0.26–4.9 | 0.88 |
| HTR3A: rs1062613 T/T | 1/2 (50%) | 10/53 (19%) | 0.78 | 4.3 | 0.25–75 | 0.32 |
| HTR3B: AAG deletion (−) | 8/42 (19%) | 3/13 (23%) | 0.75 | 1.3 | 0.28–5.7 | 0.75 |
| HTR3B: rs1176744 T/T | 4/22 (18%) | 7/33 (21%) | 0.78 | 0.83 | 0.21–3.2 | 0.78 |
| HTR3B: rs1176744 G/G | 1/7 (14%) | 10/48 (21%) | 0.69 | 0.63 | 0.068–5.9 | 0.69 |
| HTR3C: rs6766410 A/A | 5/17 (29%) | 6/38 (16%) | 0.24 | 2.2 | 0.57–8.7 | 0.25 |
| HTR3C: rs6766410 C/C | 1/11 (9.0%) | 10/44 (23%) | 0.31 | 0.34 | 0.039–3.0 | 0.33 |
| DRD2: rs6277 C/C | 7/43 (16%) | 4/12 (33%) | 0.19 | 0.39 | 0.091–1.7 | 0.20 |
| DRD2: rs6277 T/T | 0/1 (0%) | 11/54 (20%) | 0.61 | — | — | 1.0 |
| DRD2: rs1076560 C/C | 8/24 (33%) | 3/31 (10%) | 0.030* | 4.7 | 1.1–20 | 0.039* |
| DRD2: rs1076560 A/A | 0/5 (0%) | 11/50 (22%) | 0.24 | — | — | 1.0 |
| COMT: rs4680 G/G | 6/31 (19%) | 5/24 (21%) | 0.89 | 0.91 | 0.24–3.4 | 0.89 |
| COMT: rs4680 A/A | 0/3 (0%) | 11/52 (21%) | 0.37 | — | — | 1.0 |
“Yes” is percentage of patients expressed vomiting (≧ grade 1) on the worst day during the 10 d after chemotherapy in patients with each risk factor. “No” is percentage of patients expressed vomiting (≧ grade 1) on the worst day during 10 d after chemotherapy in patients without each risk factor. * p<0.05, ** p<0.01, # p<0.1 vs. corresponding “No” [a) chi-square test, b) univariate logistic regression analysis]. NSAID, non-steroidal anti-inflammatory drug; HTR3A, serotonin 5-HT3A receptor; HTR3B, serotonin 5-HT3B receptor; HTR3C, serotonin 5-HT3C receptor; DRD2, dopamine D2 receptor; COMT, catechol-O-methyltransferase.
We also evaluated separately cisplatin- and carboplatin-based chemotherapies with emetogenicity. There were no significant differences in the frequency of anorexia, nausea, and vomiting between them (Table S1). The frequencies of anorexia, nausea, and vomiting in cisplatin-based chemotherapy according to each risk factor are shown in Table S2–S4. The ORs of C/C homozygosity of rs1062613 (HTR3A) and C/C homozygosity of rs1076560 (DRD2) were significantly higher in anorexia and nausea than those of the T allele carriers (OR, 7.0; 95% CI, 1.0–47; p<0.05) and the A allele carriers (OR, 7.0; 95% CI, 1.0–47; p<0.05). In carboplatin-based chemotherapy (Tables S5–S7), the ORs of female gender and use of NSAIDs were significantly higher than those of male gender (OR, 27; 95% CI, 3.7–189; p<0.01) and no use of NSAIDs (OR, 9.3; 95% CI, 1.5–58; p<0.05).
We carried out multivariate analysis, but there were no significant relationships between risk factors and the frequency of CINV (data not shown).
We investigated whether the individual variability in the expression of CINV was associated with behavioral (patient-related) factors and polymorphisms related to the appearance of CINV in Japanese subjects. Our results suggest that determining behavioral risk factors and genetic risk factors can help to inform individually based medication for treating or preventing CINV, using novel information for cancer patients who received chemotherapy containing either cisplatin or carboplatin. Especially, determining an individual’s genotype is important to predict the clinical responses to chemotherapy, whereas it is difficult to incorporate the rapidly accumulating genome information for the Japanese cancer patients because of genetic differences among races.
CINV is one of the more debilitating side effects of chemotherapy in cancer patients. Hence, many treatment guidelines recommend the use of appropriate antiemetic agents based on the time of onset of emesis as well as the emetic risk of chemotherapeutic drugs.3) Even though the guidelines for the administration of such antiemetic treatments for CINV are generally followed, these treatments have not completely prevented CINV in certain patient populations. The predicted risk factors in the present study for CINV can be an aid to clinical decision-making and assist clinicians to rationalize antiemetic use with their patients. Depending on the relationship between CINV-related risk factors and a tailored antiemetic treatment, patients with the identified risk factors have a high risk of CINV, and may be candidates for future clinical trials attempting to improve anti-CINV results in them undergoing antiemetic treatments. Thus, these support the idea of establishing individualized supportive therapies, e.g. some additional prophylactic antiemetics and/or other medications such as anxiolytics and antidepressants for patients with moderately and highly emetogenic chemotherapy, and contribute to the development of more effective and safer chemotherapies.19) Our results could help to determine individual-based medications for treating and/or preventing CINV in Japanese cancer patients.
Several behavioral risk factors for CINV have already been identified, such as being female and younger than 50 years of age.4,20–23) In addition to chemotherapy, radiation therapy,3,13) opioid therapy,14,15) and concomitant use of NSAIDs have also been identified as factors that increase the incidence and intensity of CINV, as they are highly likely to cause gastrointestinal problems. Factors that predict nausea, vomiting, and anxiety have been reported to be risk factors for CINV as well.24–26) These previous studies have only investigated the incidence of CINV for 7 d after chemotherapy and evaluated using the visual analogue scale (VAS) or self-report questionnaire of patients themselves. Tamura et al. demonstrated that analysis of the frequency of CINV for over 7 d was needed considering patients digestive symptoms.19) The current study is the first to investigate association between each risk factor and the frequency of CINV for long periods using CTCAE. We evaluated anorexia, nausea, and vomiting independently because risk factors of expression, aggravation, and/or affecting QOL for anorexia, nausea, and vomiting are different.27,28) In fact, any of risk factors did not match among symptoms in the present analysis.
We found that the frequency of anorexia/vomiting was significantly higher in patients who were female gender, prone to motion sickness, and use of NSAIDs. Although a number of literatures have shown that age was significant risk for the incidence of CINV,29–31) several authors have been reported that age was not associated with delayed CINV or additional antiemetics.24,28,32) Our preliminary analysis indicated that the intensity of anorexia/nausea in patients under 70 years of age was significantly higher than that in patients over 70 years of age (data not shown). We will need to assess the association between behavioral factors and intensity of CINV to address this issue further in detail. Our data, together with previous reports,3,4,14,15,20–23) suggest that female gender, use of NSAIDs, susceptibility to motion sickness, and anxiety owing to their chemotherapy are risk factors for CINV. Although multivariate logistic regression analysis can identify robust factors, the multivariate model failed to converge due to insufficient sample size and the potential interactions among the factors (e.g. multicollinearity), thus further association analyses of risk factors related to CINV using large sample cohort are needed.
Analysis of gene polymorphisms such as those in the serotonin 5-HT3A receptor,7) 5-HT3B receptor,6) 5-HT3C receptor,8) cytochrome P-450 2D6,33) and P-glycoprotein34) for the purpose of investigating whether genetic factors influence the efficacy of the antiemetic agents used for CINV revealed that certain gene polymorphisms were associated with nausea and vomiting. However, there have been no studies to investigate the association between gene polymorphisms and the efficacy of antiemetic therapies on CINV in Japanese cancer patients. Since gene polymorphisms are differently distributed depending on race,35) we analyzed the association between various gene polymorphisms and the frequency of CINV in a Japanese patient population. We found an increase in the frequency of nausea and vomiting in patients with only C/C homozygosity of rs1076560 (DRD2). Several reports suggested that A allele carriers of rs1076560 have a lower ability to synthesize dopamine D2 receptor,9,10,12) suggesting that the dopamine signaling via dopamine D2 receptor located in the chemoreceptor trigger zone (CTZ) might be attenuated in A allele carriers. These reports are consistent with our results and suggest that C/C homozygotes of rs1076560 is risk factor for CINV.
Kaiser et al.7) reported no association between the 21 polymorphisms of HTR3A and the efficacy of antiemetic agents against CINV; this is consistent with our results in Japanese cancer patients. In the present analysis, however, our results of rs6766410 (HTR3C) and rs4680 (COMT) were inconsistent with the previous findings, which CINV occurred in 50% of the patients with C/C homozygosity of rs6766410 (HTR3C), compared to 19% of those with A/A homozygosity and 22% of those with A/C heterozygosity.8) Further, COMT activity in G/G homozygotes on rs4680 was approximately 3 to 4 times higher than that in A/A homozygotes.11) Our preliminary analysis indicated that the intensity of anorexia in the C allele carriers of rs6766410 (HTR3C) and in patients with G/G homozygosity of rs4680 (COMT) tended to higher than that in A allele carriers (data not shown). Our study showed that polymorphisms in all the above genes are suggested to be risk factors for CINV; however, further studies are necessary to determine the associations more clearly.
We evaluated separately cisplatin- and carboplatin-based chemotherapies with emetogenicity. There was no significant difference in the frequency of CINV between them. Although we found an increase in the frequency of anorexia in patients with only C/C homozygosity of rs1062613 (HTR3A) in cisplatin-based chemotherapy, additional studies are needed to determine the risk factors with more samples to carry out multivariate analysis and clarify risk factors between regimens.
In conclusion, we identified the individual variability in susceptibility to CINV related to behavioral factors and genetic factors in Japanese cancer patients. These support the idea of establishing individualized supportive therapies (some additional prophylactic antiemetics and/or other medications such as anxiolytics and antidepressants) for patients with risk factors of CINV. Our study theoretically contributes to increasing the safety of chemotherapy with supportive therapy to prevent CINV and increase the QOL of cancer patients. In the future, we plan to analyze the patients with the same condition, e.g. type of cancer and examine combinations of potential risk factors in order to better rank their prognostic values with larger sample cohort. Such investigations support the idea of establishing individualized supportive therapies for CINV, and contribute to the development of more effective and safer chemotherapies.
This study was supported by the ‘Academic Frontier’ Project for Private Universities (2007–2011); Grants-in-Aid for Scientific Research C (24590219, 16K08421) from the Ministry of Education, Culture, Sports, Science and Technology (MEXT) of Japan, and The Adaptable and Seamless Technology Transfer Program through Target-driven R&D, Japan Science and Technology Agency (AS251Z03018). We would like to thank Ms. Fumika Matsuoka, Ms. Misato Ota, Ms. Kaori Kato, Ms. Chihiro Murosaki, and all staff members of Faculty of Pharmacy, Meijo University that were involved in this study.
Y. Hasegawa had received Grant from Organization Astellas, Chugai, Eli Lilly, Boehringer Ingelheim, Taiho Phamaceutical, Ono Phamaceutical, and speaker’s bureau for Organization Boehringer Ingelheim, MSD, Eli Lilly, and AstraZeneca.
The online version of this article contains supplementary materials.
