To the Editor:

We were intrigued by the article titled “A Metabolic Map and Artificial Intelligence-Aided Identification of Nasopharyngeal Carcinoma via a Single-Cell Raman Platform,” authored by Xu et al. [1]. In their study, the authors employed a single-cell Raman platform to create a metabolic map encompassing the following: (A) nasopharyngeal carcinoma (NPC) cell lines: They analyzed seven “authenticated” NPC cell lines and one immortalized nasopharyngeal epithelial cell line; (B) Tissue samples: The study included six nasopharyngeal mucosa tissues and seven NPC tissue samples. Through confocal Raman spectroscopic measurements and imaging, the researchers explored altered metabolic processes involving nucleic acids, amino acids, lipids, and sugars. Additionally, they investigated metabolomic profiles using ultra-performance liquid chromatography-tandem mass spectrometry (UPLC-MS/MS). Notably, the authors proposed that confocal Raman spectroscopy, along with imaging, serves as a highly sensitive diagnostic tool for classifying NPC cells and tissue. While the study provides a wealth of experimental data, it is essential to acknowledge that more than half of the cell lines used in this research have been misidentified as NPC cells, as reported in the Register of Misidentified Cell Lines (version 12, released on January 16, 2023) (Table 1). Out of the seven cell lines analyzed, five exhibit a high degree of genetic overlap with the HeLa short tandem repeat (STR) profile, specifically the CNE1, CNE2, SUNE1, 6-10B, and 5-8F lines. It is acknowledged that the CNE1, CNE2, and SUNE1 lines were initially derived from nasopharyngeal carcinoma. However, due to prolonged culturing and suboptimal laboratory practices, these lines have become contaminated with HeLa cells. Furthermore, the SUNE-1 line serves as the progenitor for the 6-10B and 5-8F lines, which, regrettably, also show genetic similarities to the HeLa STR profile.

Table 1 Summary of nasopharyngeal carcinoma cell lines used.
Full size table

The issue of cross-contamination in nasopharyngeal carcinoma cell lines has been a recognized concern since 2008 and has been consistently emphasized in subsequent literature [2, 3]. The latest release from the International Cell Line Authentication Committee (ICLAC), their 12th edition, identifies eight “NPC cell lines” as compromised by cross-contamination [4]. Despite clear genetic evidence indicating the presence of HeLa cell contamination, many research groups focusing on NPC have persisted in utilizing these compromised cell lines to date.

Improving cell line selection and reporting is crucial for research accuracy. However, the prevalence of publications with incorrect cell line data is concerning, as it leads to their continued use and risks the validity of research [5, 6]. Addressing this issue is essential for ensuring reliable scientific outcomes. It is important to establish stringent verification processes and foster a culture of transparency in research practices. Mandatory methods such as short tandem repeat (STR) profiling and DNA fingerprinting are needed to prevent cell line misidentification and safeguard scientific integrity.

The results presented show no significant differences between HeLa-contaminated cell lines and uncontaminated ones. Therefore, the method used does not appear to distinguish between tumor cell lines of various epithelial origins. However, in the study by Xu et al., the presence of HeLa contamination in NPC cell lines does not necessarily invalidate the research findings. Along with the contaminated lines, two uncontaminated NPC cell lines, HK1 (also known as NPC/HK1) and C666-1, were also used. The C666-1 line holds particular importance in NPC research as it consistently carries EBV, reflecting the viral patterns found in most primary NPC biopsies from Chinese patients [7]. The utilization of actual tissue samples further bolsters the study’s validity. Nevertheless, accurate identification of cell lines is crucial to prevent data misinterpretation.