Comparative Analysis of Physical and Polymer Characteristics of Microplastics Detected in Human Colorectal Cancer Samples From the United States and Malaysia.

1
Introduction
Microplastics, defined as plastic particles smaller than 5 mm in size, have gained widespread attention due to their ubiquitous presence in various environmental compartments, including marine and freshwater ecosystems, as well as their potential to enter the food chain and human tissues [1, 2]. Colorectal cancer is a significant global health concern, and emerging research suggests that environmental factors, including potential exposure to microplastics, may play a role in carcinogenesis [3]. Due to their tiny size, microplastics can penetrate biological barriers and enter various body compartments including organs and probably cancer tissues too [4]. Recent studies have detected microplastics in various body systems such as gastrointestinal system, respiratory system, blood immune system, brain and nervous system, endocrine system, and reproductive system [5].
Our previously published study has detected microplastics in colectomy samples and later replicated by other colleagues; however, available studies are limited by small sample sizes and population type. The current study aimed to expand data collection using human colorectal cancer samples from the Malaysian population and also including population samples from the United States but collected at two different time points, with the American samples collected from 1993 to 1999 and the Malaysian samples being more recent, from 2023 to 2024. The comparative analysis may provide insights on similarities and differences in physical and polymer characteristics of microplastics between the two populations.

2
Materials and Methods
Inclusion criteria were human tissue samples of colorectal cancer obtained from two populations, that is, United States (University of Washington, Seattle) and Malaysia (Hospital Pakar Universiti Sains Malaysia, Kota Bharu). Exclusion criteria were non‐colorectal cancer samples or colectomy samples of other diagnoses, including inflammatory bowel disease or intestinal obstruction. A total of 25 colectomy tissue samples from Malaysia were collected from surgeries conducted between July 14, 2023, and July 13, 2024. A minimum of 1 cm of fresh colorectal cancer tissue from each patient was harvested by a pathologist and preserved in filtered absolute ethanol at room temperature. In contrast, the American archived samples were obtained from surgeries conducted between August 23, 1993, and July 15, 1999. A total of 40 formalin‐fixed, paraffin‐embedded (FFPE) colorectal cancer tissue blocks were retrieved from the university's biorepository. However, 15 samples were excluded due to insufficient tissue size for microplastic analysis, leaving 25 samples suitable for further examination.
2.1
Quality Control of Environmental Contamination of Microplastics
For any experiments involving tissue samples, a rigorous quality control and assurance protocol would be implemented to minimize potential contamination during collection, transportation, and processing of samples [6]. To reduce plastic usage, all samples were collected and stored in either glass jars or aluminum containers. Each sample was assigned a unique code number to ensure confidentiality. Throughout sample handling, researchers wore nitrile gloves and 100% cotton lab coats. To maintain the purity of reagents used, absolute ethanol, 10% potassium hydroxide, and deionized water were initially filtered through a 1.2 μm GF/C glass microfiber membrane. Furthermore, all glassware and equipment underwent at least three rinses with filtered deionized water. Prior to commencing any experiments, all laboratory bench surfaces were wiped with 80% ethanol. To prevent the introduction of uncontrolled airborne particles, analytical steps such as weighing and digestion were performed in a laminar flow hood [7]. To account for potential contamination, a procedural blank was prepared during each experimental step. This involved placing a clean membrane in an open glass petri dish, which was then covered with foil to prevent external contamination.

2.2
Extraction of Microplastics From the American Samples
Prior to deparaffinization, human FFPE tissue blocks were first cut according to the shape of samples and heated at 45°C for 30 min to slowly melt the outer layer of paraffin. The deparaffinization process was modified based on the protocols outlined in [8, 9]. First, to remove the paraffin that penetrated into the tissue, initial washing was performed with 100% xylene for 10 min, in two cycles, followed by washing with 100% ethanol for 10 min, also in two cycles, to remove the remaining xylene.

2.3
Tissue Sample Digestion
All tissues from two sites were weighed using an analytical balance before being digested with 10% potassium hydroxide at 40°C for 7–10 h at a ratio of 1:8 (w/v). This procedure was conducted with modifications to the original method [7, 10]. The digestates were then diluted with filtered deionized water before proceeding to the filtration process by passing through a 1.2‐μm glass microfiber membrane and dried in a glass desiccator.

2.4
Microplastic Extraction and Analysis
Extracted microplastic particles were sorted manually from the membrane using sharp stainless steel forceps under a dissecting microscope (Carl‐Zeiss Stemi 508, China) with 0.8–5.6× magnifications coupled to a digital camera (Axiocam 208 color). A glass chamber was used throughout the sorting process to minimize potential airborne contamination. These sorted microplastics were categorized into shape and color. Additionally, the microplastics were classified into four size categories, that is, Category I (< 300 μm), II (301–500 μm), III (501–1000 μm), and IV (1001–5000 μm) [11]. Microplastics identification and quantification were performed using micro‐Fourier transform infrared spectroscopy (μFTIR) in attenuated total reflectance (ATR) mode of analysis (LUMOS Bruker, USA), operating within the vibrational wavelength range of 4000–400 cm−1. Collected FTIR spectra were processed by OPUS software and identified by comparing to available databases in the library (Bruker Optics, USA). Surface morphology of microplastics was characterized by scanning electron microscope (SEM; JOEL JSM‐6360LA, Japan) to observe potential degradation features such as cracks, pit holes, biofilms, and adhering particles [12]. Prior to imaging, a conductive gold coating was sputtered onto the aluminium stub containing attached microplastics prior to the imaging process. SEM analysis was conducted at an optimized accelerating voltage of 15 kV and a detector working distance of 14–15 mm.

2.5
Data Analysis
The datasets were analyzed using OriginPro Version 8.1 (OriginLab, USA) for FTIR spectra, R Statistical Analysis Version 3.3.0 (RStudio, Austria) and XLSTAT Version 2019 (Addinsoft, USA). Shapiro–Wilk technique was used to assess the normality distribution of the datasets and further compared using the Mann–Whitney U test [13]. To identify variations in microplastic concentrations from both the American and Malaysian samples, the Kruskal–Wallis test with Bonferroni correction (significance level: p < 0.05) was utilized. Common multivariate analysis of principal component analysis (PCA) was utilized to construct a biplot illustrating the relationship between physical characteristics of microplastics (color, shape, and categories of sizes) and concentration of microplastics. Sampling adequacy for factor analysis was assessed through the computation of Kaiser–Meyer–Olkin (KMO) measures, and the validity of factor analysis was tested using Bartlett's test of sphericity.

2.6
Ethics Requirements
Ethical approval for the study was obtained from the National Medical Research Register of Malaysia (reference ID: NMRR ID‐23‐00196‐WJ0 (IIR)). All data collected were anonymized to ensure participant confidentiality. The use of archival samples complied with ethical guidelines and institutional regulations applicable at the time of data collection.

3
Results
A total of 50 samples of colorectal cancer tissues were analyzed, including 25 American samples and 25 Malaysian samples. Microplastics were found in all 25 American samples, with an average count of 25.00 ± 40.57 items of microplastic per gram tissue (item/g) (mean weight = 0.30 g), and likewise in all 25 Malaysian samples with an average count of 32.22 ± 48.14 item/g (mean weight = 0.53 g). The predominant color observed in both samples was blue: 44.97% from Malaysian samples and 44.58% in the American samples (Figure 1A,B). However, American samples displayed a different color distribution with the absence of yellow‐colored microplastics, but present in the Malaysian samples. Conversely, the American samples have brown‐colored microplastics but not in the Malaysian samples. Other colors found in both population samples included transparent, black, red, blue, green, and purple (Figure 2).
Additionally, similar shapes of microplastics were identified in both population samples, including fiber, fragment, and film (Figures 1C,D and 2), but fiber was the most abundant, likewise in both groups of samples. The size of microplastic particles was varied between 40 and 3011 μm in the Malaysian samples and 110 and 3165 μm in the American samples. Size Category I (< 300 μm) and III (501–1000 μm) were the most frequent in Malaysian samples, whereas Category IV with 1001–5000 μm was the most prevalent in the American samples (Figure 1E,F). Surface roughness was observed from SEM in both groups of samples (Figure 3). The microplastic particles from Malaysian samples seem to maintain their original shape with minimal cracks, contrasting with the American particles that exhibited edge ruptures at 50 μm (Figure 3C,D). Upon closer examination at 10 μm, it became evident that the American particles displayed a significantly higher degree of brittleness and appeared to be fragmented, which could lead to the generation of additional microplastic particles (Figure 3F).
Using μFTIR, the following polymer types were identified from samples of both populations: polyethylene (PE), polypropylene (PP), polyamide (PA), and polycarbonate (PC). The functional groups attributed to PE were detected at 2916 cm−1 (CH2 asymmetric stretching), 2849 cm−1 (CH2 symmetric stretching), 1464 cm−1 (CH bending deformation), and 719 cm−1 (CH rocking deformation) (Figure 4). Meanwhile, the peak assignments for PP polymers were shown at 2920 and 2850 cm−1 (CH stretchings), 1458 cm−1 (CH2 bending), 1376 cm−1 (CH3 bending), 975 cm−1 (CH3 rocking), and 810 cm−1 (CH2 rocking). Polyamide (PA) exhibited a distinctive and sharp peak at 3350 cm−1 corresponding to NH stretching, followed by a bending vibration at 1450 cm−1. Additionally, strong CN and C=O carbonyl stretching bands were observed at 1100 and 1750 cm−1 wavelengths, respectively, within the fingerprint region of the PA spectrum. The FTIR spectrum of PC displayed characteristic peaks at 2966 cm−1 (CH stretching), 1768 cm−1 (C=O stretching), 1186 cm−1 and 1158 cm−1 (C–O stretching), along with aromatic ring vibrations at 1503 cm−1 and 1409 cm−1 (aromatic C=C stretching), 1013 cm−1 (aromatic bending) and 828 cm−1 (C–H out‐of‐plane loop bending). Notably, polyethylene terephthalate (PET) was detected exclusively in the Malaysian samples, with functional group characteristic peaks at 1713 cm−1 (C=O stretching), 1242 cm−1 and 1241 cm−1 (C–O stretching), and 720 cm−1 (aromatic C–H oop bending). Conversely, acrylonitrile butadiene styrene (ABS) was identified only in the American samples, indicated by CH stretching at 2922 cm−1, accompanied by distinctive peaks at 1602 and 1494 cm−1, and 759 cm−1 (CH oop bending), confirming the presence of aromatic ring structures.
PCA of microplastics abundance and physical characteristics is illustrated in Figure 5. The X‐axis, denoted as F1, accounted for 36.06% of the data variability, whereas the Y‐axis, denoted as F2, represented 18.64% of the variance. These two components (F1 and F2) represented 54.70% of the total variance in the given data with eigenvalues of 3.245 and 1.678, respectively. The KMO correlation coefficient was 0.568, and the Bartlett's sphericity test was significant (p < 0.0001). Based on PCA, microplastic abundance was distinctly different between Malaysian and American samples. There was a strong correlation between two colors (blue, red) and categories of sizes (I, II, and III) in the Malaysian samples. Factor loading 1 (FL) revealed that fiber was the most frequent component (0.889), followed by blue‐colored microplastics (0.758) and category size I (0.757) (Table 1). Conversely, the characteristics of microplastics had negative loadings towards American samples, indicating lower correlations.

4
Discussion
The first notable finding is the ubiquitous detection of microplastics in all colorectal cancer tissues of both the American and Malaysian samples. Our group was the first to report the ubiquitous presence of microplastics in the bowels and later replicated by another group [3, 7]. The current study examined the largest colon samples by far, from two different populations, and the samples were obtained from two different historical time points. There are a number of significant observations by comparative analysis of the two populations described below.
First, microplastics were significantly more abundant in the Malaysian versus American samples (mean = 32.2 ± 48.14 particles/g vs. 25.9 ± 40.57 particles/g, respectively). The abundance found in bowels is relatively higher than microplastics observed in testis (11.60 particles/g) [14], lower limb joints (5.24 particles/g) [15], liver (4.6 particles/g) [16], lung (2.84 particles/g) [17], and placenta (2.70 particles/g) [18]. One plausible explanation for this elevated burden of microplastics among Malaysians is dietary exposure, particularly seafood. In 2020, Malaysia reported a per capita seafood consumption of 53.33 kg annually, more than twice that of the United States (22.45 kg/year) [19, 20]. Recent estimates suggest that Malaysian adults may ingest up to 478.16 microplastic particles per year through fish consumption alone [21]. Another explanation may be historical differences in environmental plastic contamination. Microplastic exposure in the 1990s was likely lower than the present levels, as the widespread use of plastic products in the United States, such as plastic bags, did not occur until around 1979 [22].
Second, longer microplastic particles were observed in the American versus Malaysian samples. The exact reasons are unknown, but the age of samples might play a role. In the 1990s' United States, plastics were likely larger and less degradable because of early polymer technology [23]. The early form of plastics was typically made from traditional petrochemical‐based polymers, such as polyethylene and polypropylene [24]. Small microplastics (< 300 μm) were more abundant in the Malaysian cancer samples, and smaller microplastic particles (1–300 μm) have been shown to easily penetrate biological barriers, including the gut wall, and translocate into the systemic circulation and smaller vessels [25, 26]. Vasculatures in tumors are often erratic and porous, allowing easy penetration and accumulation of microplastics [27, 28, 29].
Third, microplastics from the American samples displayed significant signs of fragmentation and more surface degradation compared to those in the Malaysian samples (Figure 3). The above findings may be attributed to the aging process of microplastics and from chronic exposure of microplastics to environmental factors [25]. Microplastics have the ability to accumulate oxidizing radicals on their surfaces [30], which may change the chemical and structural properties of microplastics, such as surface cracking, chemical leaching, altered crystallinity, increased hydrophilicity, and particle aggregation [31]. In the bowels, microplastics may undergo further degradation due to exposure to fluids of extreme pH and also by a myriad of gut enzymes [7]. In addition, the FFPE in American samples required deparaffinization using xylene, a nonpolar solvent that may cause brittleness of microplastics [32]. Repeated solvent washes during FFPE processing could also result in the loss of smaller or loosely bound microplastic particles.
Fourth, certain polymers were uniquely identified in each population sample, suggesting population‐specific exposure pathways influenced by regional plastic usage patterns, lifestyle practices, and environmental factors. Notably, ABS was detected exclusively in the American samples. In the past, ABS has been widely used in the production of children's toys, household appliances, and electronic casings due to its high durability and ease of molding [33, 34].
Fifth, the eigenvalue of microplastics found in the Malaysian cancer samples closely aligned with the eigenvalue of fibers (Figure 5). This observation may be linked to dietary habits, as Malaysians habitually consumed seafood, and the local studies have reported a high prevalence of microplastic fibers ranging from 500 to 1500 μm in length, extracted from four major commercial fish in Malaysia, namely, yellowtail scad (
Atule mate
), bluespot mullet (
Crenimugil seheli
), fringescale sardinella (
Sardinella fimbriata
), and short mackerel (
Rastrelliger brachysoma
) [35]. Another study further identified microplastics in
A. mate
,
R. brachysoma
, round scad (
Decapterus punctatus
), and silver pomfret (
Pampus argenteus
), including particles as small as 0.04 mm [36].
Several limitations also should be acknowledged in the present study. First, it is limited by insufficient clinical information, especially the age of participants and including dietary habits. As chronic exposure to environmental risk factors increases with time, therefore the absence of age information prevented any meaningful comparison of microplastics exposure duration (over years of life) between the two cohorts. Second, although our study does not provide a confirmatory link between microplastics exposure and colorectal cancer, it provides a basis and allows us to test further hypotheses on the carcinogenic roles of microplastics for future investigation. Third, there is a temporal discrepancy between the American and Malaysian samples, as the former were collected in the 1990s. Although the possibility of contamination during the historical collection of American samples cannot be entirely ruled out, however, the many similarities between both sample sets demonstrated in our analyses indicate that the contamination was likely minimal. Fourth, the use of xylene for deparaffinizing FFPE tissues along with the multiple solvent washes may have affected the integrity of microplastics, particularly those with lower chemical resistance (e.g., polystyrene), by inducing fragmentation, brittleness, or partial dissolution. Our future studies will consider experimenting with laboratory effects of FFPE deparaffinization and washing on microplastics, including exploring alternative methods to assess and minimize solvent‐induced degradation effects. Enhanced recovery techniques and analysis of wash fractions will also be considered to improve the accuracy of microplastic quantification.
As a conclusion, microplastics are ubiquitously detected in all American and Malaysian samples of colorectal cancer tissues despite the different sampling time points. There are similarities but also distinct characteristics between the two population samples, including differences in abundance, roughness, size/length, and polymer type.

Conflicts of Interest
The authors declare no conflicts of interest.