‏Exposure to phthalate-containing prescription drugs and the risk of colorectal adenocarcinoma: a Danish nationwide case-control study

Abstract

Background

Some drug products contain phthalates as excipients and in vitro studies have demonstrated that phthalates interfere with cellular mechanisms triggering colorectal cancer development. We therefore examined the association between cumulative phthalate exposure from drug products and risk of colorectal adenocarcinomas.

Methods

We used the Danish Cancer Registry to identify all patients with incident colorectal adenocarcinoma from 2008–2015 (n=25,814). Each cancer case was matched to ten population controls. Linking information from Danish registers, we quantified cumulative phthalate exposure to the ortho-phthalates diethyl phthalate (DEP) and dibutyl phthalate (DBP) as well as enteric phthalate polymers from orally administered drugs. The association between cumulative phthalate exposure and colorectal cancer was estimated using conditional logistic regression.

Results

Cumulative exposure to ortho-phthalates exceeding 500 mg was associated with lower odds of colorectal cancer diagnosis (OR_adj 0.89; 95% CI: 0.81–0.96). Similar associations were observed for all DEP exposure exceeding 500 mg. Subgroup analysis excluding NSAID-users, ortho-phthalate exposure was positively association with colorectal cancer (OR_adj=1.26, 95%CI: 1.05–1.51).

Conclusion

We found an apparent overall protective effect of cumulative phthalate exposure from drug excipients for colorectal adenocarcinoma. Omitting NSAID users reversed the signal, and suggested a slightly increased risk associated with high cumulative ortho-phthalate exposure.

Introduction

The widespread use of phthalates as softeners in plastic products has gained increasing attention since phthalates are suspected endocrine disruptors and reproductive toxins ¹. Although the carcinogenic properties of phthalates are still uncharacterized, some epidemiologic studies have associated ortho-phthalate exposure with incident breast cancer in humans ^2,3.

Ortho-phthalates are known to stimulate the epithelial-to-mesenchymal transition necessary for cancer invasion and metastasis, mediated by histone deacetylase 6 ⁴. In vitro studies support a role of phthalates in mechanisms of colon carcinogenesis. Additionally, a dose-dependent effect of ortho-phthalates on aryl hydrocarbon receptor (AhR) function has been suggested. The AhR is a transcription factor that plays an important role in cell proliferation, differentiation and in tumor development ⁵.

Some phthalates are used as excipients in pharmaceuticals. Their water resistant and acid stable properties are utilized in the pharmaceutical production of sustained or delayed release preparations ⁶. Among users of phthalate containing drugs, an up to 50-fold increased urinary concentration of phthalate metabolites was demonstrated compared with non-users ⁷. The European Medicines Agency (EMA) and the U.S. Food and Drug Administration (FDA) are aware of the potential harmful phthalate exposure from pharmaceutical preparations, and both agencies have published guidelines on limiting pharmaceutical phthalate exposure from orally administered drug products ^8,9. Recent population-based Danish data suggest that exposure to phthalates from drugs is not negligible, with a median dibutyl phthalate (DBP) exposure from prescription drugs that remained above European regulatory limit of exposure ^10,11.

This led us to consider if the phthalate exposure from orally administered drugs carries an excess risk of developing colorectal adenocarcinomas. We therefore performed a Danish nationwide registry-based case-control study to examine the association between cumulative pharmaceutical phthalate exposure and risk of colorectal adenocarcinoma.

Methods and materials

The study was conducted according to STROBE guidelines for reporting observational studies ¹², using a nationwide case-control approach investigating the cumulative phthalate exposure from orally administered drug products among individuals diagnosed with colorectal cancer (cases) and population based cancer-free persons (controls).

Data sources

Virtually all medical care in Denmark is furnished by the national health authorities, allowing true population-based register linkage studies covering all inhabitants of Denmark. A variety of information is captured in the Danish Registries ¹³. Data were linked by the personal identification number, a unique identifier assigned to all Danish residents since 1968 ¹⁴. We used information from five Danish nationwide registries: the Danish Cancer Registry ¹⁵, the National Prescription Registry ¹⁶, the National Patient Register ¹⁷, Registers in Statistics Denmark on educational level ¹⁸, and the Civil Registration System ¹³.

Detailed information on data sources is presented in Appendix A and codes for cancer diagnoses, drug exposures and covariates are available in Appendix B.

By linking data from an internal database maintained by the Danish Medicines Agency with the National Prescription Registry, we could quantify phthalate exposure on the level of the individual. The internal Danish Medicines Agency database provided detailed information on phthalate content per tablet or capsule, in specific marketed drug-products. This database records information on type and amount of all excipients in drugs with Danish marketing permission from 2004 onwards. All changes in quantitative compositions of excipients are recorded. Each specific drug product can be identified by the Nordic product code (VNR) and drugs are classified according to the Anatomic Therapeutic Chemical (ATC) index, developed by the World Health Organization. Only orally administered drugs were included.

Sampling of cases and controls

From the Danish Cancer Registry, we identified all patients (cases) in Denmark with a first-time diagnosis of colorectal cancer (ICD10: C18–20) in the period 2008–2015. The case population was restricted to individuals with histologically verified adenocarcinoma. The date of the cancer diagnosis was used as the index date. Exclusion criteria were age outside the range 18–85 years at index date and any residency outside Denmark within 10 years prior to index date. We further excluded cases with a history of other cancers (except non-melanoma skin cancer) as well as individuals with disorders associated with an increased risk of colorectal cancer: hereditary nonpolyposis colon cancer, familial adenomatous polyposis, and inflammatory bowel disease (IBD); defined either through a diagnosis or use of anti-inflammatory drugs specific to IBD, ATC-group A07E). Lastly, patients with total colectomy were excluded.

Controls were selected using risk set sampling, applying the same exclusion criteria as for cases. For each case, we selected ten cancer-free controls among all Danish residents of the same gender and birth year. Controls were were identified through the civil registration system covering all Danish inhabitants ¹³ and they were assigned an index date identical to that of the corresponding case. Subjects were eligible for being sampled as controls before they became cases.

Exposure definition

Exposure to ortho-phthalates or enteric phthalate polymers was quantified by cumulative exposure during the period 2004–2015. We did this by linking information on pack size and phthalate amount per pill or tablet for all dispensed prescriptions by all subjects included in our studies. Ever-exposed designates those, who had ≥1 phthalate containing product dispensed.

Exposure was also characterized by specific phthalate: ortho-phthalates by diethyl phthalate (DEP) and dibutyl phthalate (DBP) and the individual enteric phthalate polymers cellulose acetate phthalate (CAP), Hydroxypropyl methylcellulose phthalate (HPMCP) and polyvinyl acetate phthalate (PVAP). Exposure categories were somewhat arbitrarily. We aimed to separate the risk among those exposed to low amounts of phthalates from those exposed to larger amounts, and still provide sufficient number of subjects within each category to produce meaningful estimates.

High exposure to any ortho-phthalate was defined as >500 mg of cumulative phthalate exposure, intermediate exposure was defined as 250–499 mg of cumulative exposure, and low exposure was defined as <250 mg of cumulative exposure over the study period. For enteric phthalate polymers, the corresponding categories were >10,000 mg, 5,000–9,999 mg, and <4,999 mg of cumulative exposure. Never-exposed groups were defined for ortho-phthalates and enteric phthalate polymers, respectively.

Confounding variables

The following potential confounders were identified and incorporated as dichotomic variables in the analyses: a) Use of drugs known or suspected to modify the risk of colorectal adenocarcinoma including use of statins, antidepressants, and hormone replacement therapy, prior diagnoses of diabetes, chronic obstructive pulmonary disease (COPD), and alcohol-related disease; and c) highest achieved education (as a crude measure of socioeconomic status). Dichotomization was determined by classifying patients who redeemed at least one of the prescription drugs or receiving at least one of the ICD-10 codes during the study period as being exposed or diagnosed. Use of low-dose aspirin (ASA) and nonsteroidal anti-inflammatory drugs (NSAIDs) were incorporated as categorical variables in the analyses based on number of redeemed prescriptions. Low-dose ASA was categorized in 0, 1–15 or >15 redeemed prescriptions and NSAIDs were categorized in 0, 1–5 and >5 redeemed prescriptions. As in the assessment of drug exposure, we disregarded the period one year prior to the index date in the identification of confounder status (ICD-10 codes and ATC-codes are listed in Appendix B).

Main analysis

The analysis followed a conventional matched case-control approach. We tabulated the frequency and proportion of cases and controls within categories of the exposures and covariates. We used conditional logistic regression to estimate ORs for colorectal adenocarcinoma associated with high exposure to any ortho-phthalate, DEP, DBP or to any enteric phthalate polymer, adjusting for potential confounders. We performed dose-response analyses using above mentioned pre-defined exposure groups. In all analyses exposure to any phthalate was compared with never-exposure (reference category). In all exposure calculations, we disregarded prescriptions redeemed within one year prior to the index date. This was done to reduce the possibility of reverse causation, while also judging that such recent exposure is unlikely to affect cancer development ¹⁹.

Pre-planned sensitivity and sub-analyses

We examined heterogeneity of phthalate/colorectal adenocarcinoma associations within strata of gender, age, year of sampling, stage of disease, topography and no history of diabetes or alcohol abuse. Further we performed an analysis excluding lithium-treated patients and patients treated with NSAIDs or low-dose ASA to see if extensive representation of these specific drugs influenced our results. This was done because DBP exposure is mainly driven by lithium products ¹¹ and as long-term exposure to NSAIDs and low-dose ASA is known to reduce the risk of colorectal cancers ²⁰.

Other

The study was approved by the Danish Data Protection Agency. According to Danish law, studies based solely on register data do not require approval from an ethics review board ²¹.

Results

We identified 25,814 colorectal cancer cases with histologically verified colorectal adenocarcinoma eligible for inclusion (Figure 1). To these cases were matched 258,140 cancer-free population controls. Most of the included cases and controls were above 70 years of age (51%) and 56 % were men. For ortho-phthalate exposure, drug use, comorbidities and educational level the cases and controls were balanced. Baseline characteristics of cases and controls are shown in Table 1.

Figure 1:

Flowchart displaying selection of cases

Table 1:

Characteristics of cases and their matched controls

	Cases	Controls
All	(n=25,814)	(n=258,140)
Male gender	14,518 (56.2%)	145,180 (56.2%)
Age
Median (IQR, years)	70 (63–76)	70 (63–76)
<50 years	1,111 (4.3%)	11,110 (4.3%)
50–69 years	11,485 (44.5%)	114,850 (44.5%)
70+ years	13,218 (51.2%)	132,180 (51.2%)
Cancer location
Colon, proximal	7,442 (28.8%)	-
Colon, distal	8,197 (31.8%)	-
Colon, unknown	9,170 (35.5%)	-
Rectum	1,005 (3.9%)	-
Ortho-phthalate exposure
Never exposed	19,207 (74.4%)	189,540 (73.4%)
0–249 mg	5,511 (21.3%)	56,152 (21.8%)
250–499 mg	422 (1.6%)	4,724 (1.8%)
≥500 mg	674 (2.6%)	7,724 (3.0%)
Drug use
Low-dose aspirin	6,835 (26.5%)	69,110 (26.8%)
Non-aspirin NSAID	14,850 (57.5%)	152,492 (59.1%)
Statins	8,130 (31.5%)	79,582 (30.8%)
Estrogens	4,362 (16.9%)	47,171 (18.3%)
Antidepressants	4,300 (16.7%)	48,615 (18.8%)
Comorbidities
Diabetes	2,354 (9.1%)	20,762 (8.0%)
COPD	1,577 (6.1%)	14,371 (5.6%)
Alcohol-related diseases	1,325 (5.1%)	11,913 (4.6%)
Education
Short (7–10 years)	9,679 (37.5%)	95,104 (36.8%)
Medium (11–13 years)	10,383 (40.2%)	98,284 (38.1%)
Long (>13 years)	4,959 (19.2%)	53,232 (20.6%)
Unknown	793 (3.1%)	11,520 (4.5%)

IQR: Inter Quartile Range, NSAID’s: Non-Steroid Anti-inflammatory Drugs.

Overall, 6,607 (25.6%) of cases had used phthalate-containing orally administered drugs compared with 68,600 (26.6%) of controls. Among these 674 cases (2.6%) and 7,724 controls (2.9%) were classified as having high exposure to ortho-phthalates. This yielded an adjusted OR of 0.89 (95%CI: 0.81–0.96) for the association between high ortho-phthalate exposure and the risk of colorectal adenocarcinoma (Table 2).

Table 2:

Exposure status and cumulated exposure to any ortho-phthalate during 2004–2015. Ortho-phthalate exposure is specified to either diethyl phthalate (DEP) and dibutyl phthalate (DBP) exposure.

	Cases	Controls	Crude OR	Adjusted OR †
Ortho-phthalates exposure status
Never exposed	19,207	189,540	1.0 (ref.)	1.0 (ref.)
Ever exposed	6,607	68,600	0.95 (0.92–0.98)	0.98 (0.95–1.01)
Ever exposure to Ortho-phthalates
0 – 249 mg	5,511	56,152	0.97 (0.94–1.00)	1.00 (0.96–1.03)
250 – 499 mg	422	4,724	0.88 (0.79–0.97)	0.91 (0.82–1.01)
≥500 mg	674	7,724	0.86 (0.79–0.93)	0.89 (0.81–0.96)
DEP
0 – 249 mg	5,312	54,082	0.97 (0.94–1.00)	1.00 (0.96–1.03)
250 – 499 mg	402	4,545	0.87 (0.78–0.96)	0.90 (0.81–1.00)
≥500mg	568	6,541	0.86 (0.78–0.94)	0.88 (0.80–0.96)
DBP
0 – 249mg	394	4,124	0.96 (0.86–1.07)	1.00 (0.90–1.11)
250 – 499mg	22	283	0.81 (0.52–1.26)	0.84 (0.54–1.30)
≥500mg	114	1,253	0.90 (0.74–1.10)	0.95 (0.78–1.16)

^†Fully adjusted model included following drugs and diagnoses: statins, antidepressants, hormone replacement therapy, NSAIDs, ASA, diabetes, chronic obstructive pulmonary disease, and alcohol-related disease, highest achieved education.

Dose-dependent effect was not seen across exposure levels among those exposed to any ortho-phthalate (p=0.63, test for trend). There were no apparent associations for exposure below 500 mg. Analyses of specific phthalate exposures consistently returned estimates below the null across levels of DEP and DBP with cumulated exposure exceeding 250 mg (Table 2). Likewise, inverse associations were seen in analyses stratified by age, gender, cancer location and stage of disease.

Excluding patients diagnosed with diabetes, patients treated with lithium, patients treated with ASA products, and patients treated for alcohol abuse did not alter our findings (Table 3). However, excluding patients with either >1 or >3 redeemed NSAID prescriptions returned estimates suggesting an increased colorectal cancer risk with phthalate exposure, with ORs of 1.26 (95% CI: 1.05–1.51) and 1.21 (95% CI: 1.04–1.41), respectively (Table 3). Analysis of a possible interaction between NSAID use and phthalate ever-exposure yielded an OR of 0.92 (95%CI: 0.86–0.99), suggesting weak effect modification, whereby the OR for the phthalates/ cancer association is slightly lower in the presence of concurrent NSAID therapy.

Table 3.

Stratified analyses by age, sex, cancer localization and stage of disease. Subgroup analyses excluding or restricting to certain populations.

	Cases Exposed/unexposed	Controls Exposed/unexposed	Crude OR	Adjusted OR †
All	674 / 19,207	7,724 / 189,540	0.86 (0.79–0.93)	0.89 (0.81–0.96)
Age
<50 years	10 / 872	141 / 8,341	0.69 (0.36–1.32)	0.68 (0.35–1.31)
50–69 years	242 / 8,651	2,761 / 85,238	0.86 (0.75–0.98)	0.87 (0.76–1.00)
70+ years	422 / 9,684	4,822 / 95,961	0.87 (0.78–0.96)	0.90 (0.81–1.01)
Sex
Male	334 / 11,083	3,787 / 109,983	0.88 (0.79–0.99)	0.89 (0.79–1.00)
Female	340 / 8,124	3,937 / 79,557	0.84 (0.75–0.94)	0.89 (0.79–1.00)
Localization
Proximal	236 / 5,377	2,400 / 54,110	0.98 (0.86–1.13)	0.98 (0.85–1.13)
Distal	206 / 6,122	2,408 / 60,178	0.84 (0.73–0.98)	0.87 (0.75–1.01)
Rectum	202 / 6,971	2,575 / 67,907	0.77 (0.66–0.89)	0.81 (0.70–0.94)
Unknown	30 / 737	341 / 7,345	0.83 (0.56–1.23)	0.82 (0.55–1.23)
Stage of disease
Localized	240 / 6,649	2,781 / 66,597	0.86 (0.75–0.99)	0.84 (0.73–0.96)
Non-localized	304 / 9,170	3,554 / 89,739	0.84 (0.74–0.94)	0.91 (0.81–1.03)
Unknown	130 / 3,388	1,389 / 33,204	0.92 (0.76–1.11)	0.92 (0.76–1.11)
Other subgroups
Excluding lithium exposed	581 / 19,185	6,856 / 189,327	0.83 (0.76–0.91)	0.85 (0.78–0.93)
Excluding diabetics	561 / 17,665	6,537 / 176,156	0.85 (0.78–0.94)	0.89 (0.82–0.98)
Excluding alcohol abusers	605 / 18,337	6,885 / 182,072	0.87 (0.80–0.95)	0.91 (0.83–0.99)
Excluding NSAID exposed (≥1 prescription)	173 / 9,583	1,439 / 92,782	1.26 (1.05–1.50)	1.26 (1.05–1.51)
Excluding NSAID exposed (≥3 prescriptions)	228 / 11,861	1,939 / 114,644	1.20 (1.04–1.40)	1.21 (1.04–1.41)
Excluding ASA exposed	419 / 14,536	4,529 / 142,885	0.91 (0.82–1.01)	0.93 (0.84–1.04)
Excluding NSAID & ASA users (≥1 prescription)	477 / 16,559	5,073 / 162,174	0.92 (0.83–1.01)	0.94 (0.85–1.04)
Restricting to NSAID users	501 / 9,624	6,285 / 96,758	0.81 (0.73–0.90)	0.83 (0.75–0.92)
Restricting to ASA users	255 / 4,671	3,195 / 46,655	0.77 (0.66–0.90)	0.80 (0.68–0.94)
Restricting to diabetics	113 / 1,542	1,187 / 13,384	0.75 (0.52–1.09)	0.80 (0.55–1.17)
Restricting to alcohol abusers	69 / 870	839 / 7,468	0.78 (0.43–1.42)	0.93 (0.49–1.77)
Restricting to lithium users	581 / 19,185	6,856 / 189,327	0.83 (0.76–0.91)	0.85 (0.78–0.93)
Restricting to those sampled 2004–2009	162 / 4,409	1,771 / 43,615	0.88 (0.75–1.04)	0.90 (0.76–1.06)
Restricting to those sampled 2010–2015	512 / 14,798	5,953 / 145,925	0.85 (0.78–0.94)	0.88 (0.80–0.97)

^†Fully adjusted model included following drugs and diagnoses: statins, antidepressants, hormone replacement therapy, NSAIDs, ASA, diabetes, chronic obstructive pulmonary disease, and alcohol-related disease, highest achieved education

Results for exposure to any enteric phthalate polymer as well as exposure to the individual compounds CAP, HPMCP, and PVAP are presented in the supplementary material. Following findings were seen among those exposed to any enteric phthalate polymer; cumulative exposure between 0 – 4,999 mg yielded an adjusted OR of 0.94 (95%CI: 0.91–0.98), cumulative exposure between 5000 – 9,999 mg yielded an adjusted OR of 1.00 (95%CI: 0.89–1.12) and ≥10,000 mg resulted in an adjusted OR 1.11 (95%CI: 1.03–1.19). An equivalent pattern of was seen for HPMCP, but not for CAP and PVAP (Supplementary material; Table S1).

Changing the lag-time to either six months or two years, had no effect on our estimates when compared with 1-year lag (data not shown).

Discussion

In this study, Danish users of phthalate-containing drug products with cumulated exposure to any ortho-phthalate exceeding 500 mg across the study period appeared less likely to develop colorectal adenocarcinomas compared with non-users, with an adjusted OR of 0.89 (95% CI: 0.81–0.96). Consistency in inverse effect of DEP and DBP exposure was observed across all exposure levels exceeding 250 mg. Post-hoc analysis excluding NSAID users led to a reversal of the association, indicating a positive association between phthalate exposure and colorectal cancer risk.

Use of high-quality registries covering the entire Danish population is the main strength of our study. In addition, primary non-adherence was eliminated by use of dispensed prescriptions ²². Additionally, the influence of secondary non-adherence was reduced, because exposure was quantified in cumulative amount. Patients missing doses of their medications, often stretch the coverage of their prescription. In this way, these patients are exposed to the entire amount of phthalates from the dispensed product ²³. The principle weakness of this study is lack of information on lifestyle factors known to influence the risk of colorectal adenocarcinomas such as smoking ²⁴, alcohol consumption ²⁵, body weight ²⁶, and certain diets ²⁴.While we controlled for a number of possible confounders and covariates, we cannot exclude confounding by indication. The distribution of quantitative phthalate exposure by drug is listed in Supplementary table S2 and S3. Quantitatively, the most important contributions to the cumulative ortho-phthalate exposure were from preparations containing lithium, multienzymes, theophylline, erythromycin and diclofenac (Supplementary material; Table S2 and S3).

Our primary finding was in direct opposition to our hypothesis as data suggested a protective effect rather than representing a risk factor. We know of no plausible biological explanation supporting a protective effect of phthalates towards colorectal cancer. Subgroup analysis omitting NSAID exposed individuals eliminated the protective association and even suggested a slightly increased risk from high cumulative ortho-phthalate exposure. As this finding emerges from a post-hoc analysis, interpretation of this result should be made with caution. We believe the most plausible explanation for our primary result is substantial confounding from NSAID use. NSAID exposure is known to reduce the risk of colorectal cancers ²⁰, which could explain the protective effect observed. Ibuprofen, diclofenac, naproxen and ketoprofen were available as phthalate-containing variants throughout the study period and NSAID-products represented 14.9% of DEP-exposure and 0.9% of DBP- exposure (Supplementary material; Table S2 and S3). No phthalate containing aspirin products were available. The quantitatively most important contribution to DBP-exposure was from preparations with lithium. About 50% of lithium reimbursed in the study period contained ortho-phthalate ^10,11, but neither depression nor lithium itself appear associated with increased risk of colorectal cancer ^27,28.

Based on the currently available data the U.S. Consumer Product Safety Commission have not classified DEP and DBP as carcinogenic compounds ^29,30. Only a few epidemiological studies on the carcinogenic effect of phthalates in humans exist. Lopez et al. demonstrated elevated risk of breast cancer among Mexican women with high urinary concentrations of monoethyl phthalate (MEP), a DEP metabolite. The study compared the lowest tertile of exposure to the highest tertile and an OR of 2.20 (95%CI: 1.60–10.70) for the risk of breast cancer was found ². Holmes et al. conducted a case-control study in Alaska Native women, investigating the association between breast cancer and exposure to several environmental chemicals. The authors did not demonstrate an increased risk of breast cancer among women high urinary concentrations of MEP or monobutyl phthalate (MBP), a DBP metabolite. Urinary concentrations above the geometric mean yielded adjusted ORs of 0.55 (95%CI: 0.26–1.18) and 0.66 (95%CI: 0.32–1.39) for MEP and MBP, respectively ³. Occupational exposure to diethylhexyl phthalate containing Polyvinyl Chloride (PVC), and the risk of testicular cancer were investigated in a study by Hardell et al. Overall a harmful association between occupational PVC exposure and the risk of seminomas with an OR at 5.6 (95%CI: 1.1–196) was demonstrated ³¹. However, this study covered multiple potential harmful chemical exposures, including diethylhexyl phthalate which is not used as excipient in orally administered drugs, in consequence of PVC exposure.

Unlike our study, the study by Hardell et al. investigated occupational or environmental exposures with no quantitative control of actual exposures on the level of the individual. The studies all had small sample sizes and they were investigating the risk of cancers connected to endocrine disrupting properties of phthalates using biomonitoring and/or questionnaire data.

We had information on exposure from 2004 until one year prior to the index-date or date of receiving cancer diagnosis. We included up to 11 years of exposure during the period 2004–2015. During this period only 2.6% of cases and 3.0% of controls in this study were exposed to more than 500 mg of ortho-phthalate from prescription drugs. Compared to environmental exposure, the cumulated exposure from prescription drugs during the study period, is of modest magnitude in most of the subjects included in this study. However, some users of phthalate containing drug products are highly exposed and their exposure might be markedly higher than exposure from environmental sources ¹¹. The estimate of minimum environmental exposure to DEP and DBP exceeds 1200 mg throughout the study period. This is based on estimates of human environmental DEP-exposure between 0.0023 and 0.012 mg/kg/day or about 696 mg throughout the study period for a 70-kg person ^32–34. Likewise, DBP-exposure estimates are 0.001–0.005 mg/kg/day, or a minimum exposure about 528 mg throughout the study period for a 70-kg person ^35,36. Additionally, information on a specific source of exposure for up to 11 years is rather limited, considering that most humans are exposed to phthalates from environmental sources, throughout their entire lifetime.

Conclusion

An apparent overall slightly protective effect of cumulative phthalate exposure from drug excipients for colorectal adenocarcinoma was demonstrated. Omitting NSAID users from the analysis reversed the signal and suggested a slightly increased risk associated with high cumulative exposure to ortho-phthalates. Regardless of assumptions on causality, the signals are weak and unlikely to be of meaningful clinical relevance.

Key points.

• Some prescription drugs contain phthalates.

• Exposure may be significant

• Little is known about the carcinogenic properties of phthalates

• Omitting NSAID-use, major exposure to phthalates was associated with a slightly increased risk of colorectal adenocarcinoma.

• Findings are unlikely of clinical relevance.