Biological products are generally expressed by continuously subcultured engineered bacteria or engineered cells, such as CHO cells, Escherichia coli, Pichia pastoris and Vero cells. Despite a series of purification processes, residual DNA fragments derived from host cells (Host Cell DNA, HCD) may still exist in final products. As an inevitable process-related impurity during production, HCD poses risks including adverse immune reactions. Notably, the tumorigenic risk of residual DNA is not solely determined by fragment size; it mainly depends on whether the DNA contains complete functional oncogenic sequences and the actual exposure dose. Large DNA fragments (over 200 base pairs, bp) are more likely to carry intact gene sequences, thus presenting relatively higher potential tumorigenic and infective risks.

Regulatory authorities worldwide regard residual host cell DNA fragments in biological products as a critical indicator affecting product quality and safety, and have formulated clear regulatory requirements to control such risks.

01  Origin: Collective Concerns Over "Invisible Impurities"

Research on HCD risks dates back to the 1950s. The U.S. Army Epidemiology Board proposed banning the use of tumorigenic cells and human tumor cells for human vaccine production over concerns about the carcinogenic hazards of residual DNA. This proposal was implemented for more than 40 years, and academic disputes over the risks of residual DNA persisted across the industry for a long time.

In the 1980s and 1990s, recombinant DNA technology was put into practical use, and the first batch of biotech drugs including recombinant insulin and interferon came into being. The industry shifted its research focus from confirming the existence of risks to exploring risk thresholds. Quantitative studies on the carcinogenic risks of residual DNA were carried out based on oncogene models, and relevant quality control specifications were gradually established. In 1976, the U.S. National Institutes of Health (NIH) issued the Guidelines for Research Involving Recombinant DNA Molecules, laying a foundation for the subsequent formulation of safety standards related to HCD.

In the 1980s, the U.S. Food and Drug Administration (FDA) and the World Health Organization (WHO) released early control standards, stipulating that the HCD limit for non-oral biologics should be no more than 100 picograms (pg) per dose to realize conservative risk control through strict limits. With advances in scientific research, it was confirmed that the tumorigenic dose of complete oncogenes reaches the microgram level, which is far higher than the actual residual amount in pharmaceutical products. Accordingly, regulatory limits were optimized: the general standard was adjusted to 10 nanograms (ng) per dose, most monoclonal antibody products still adopted the limit of 100 pg per dose, and the limit for hepatitis B vaccines was further tightened to 10 pg per dose. The regulatory paradigm evolved from conservative quantity limitation to scientific quantitative risk control, which also drove the iterative upgrading of HCD detection technologies.

02 Development: Leap of HCD Detection Technologies

 The development of HCD detection technologies has undergone a progressive journey from unfeasible detection to qualitative testing, then to accurate quantification, and finally to comprehensive scientific risk assessment.

2.1 Early Stage: Hybridization Assay and Fluorescent Dye Method

· In vivo safety test: This was the core quality control method in the early stage. It could verify the overall in vivo safety of products but failed to quantify HCD or guide the optimization of production processes.

· DNA probe hybridization assay (early 1990s): It was the first specific detection method for HCD. Digoxigenin-labeled probes were combined with target DNA to produce color signals. Nevertheless, this method had obvious limitations: the cumbersome workflow took several days to complete, it could only achieve semi-quantification or relative quantification, and its reliable limit of detection was about 10 pg. Even so, it marked a pivotal breakthrough from scratch in the history of quality control and established the initial evaluation benchmark for the industry.

· Threshold assay: This method relied on the combination of biotinylated single-strand DNA binding protein (biotin-SSB) and urease-conjugated anti-single-stranded DNA monoclonal antibody with denatured DNA. Quantitative results were obtained by detecting pH changes caused by urea decomposition catalyzed by urease. However, it was susceptible to interference from small DNA fragments and suffered signal inhibition when testing high-concentration samples, and was eventually phased out.

· PicoGreen fluorescent dye method: The fluorescent dye binds to double-stranded DNA (dsDNA) and emits fluorescence under excitation at a specific wavelength, and the fluorescence intensity is positively correlated with DNA concentration, enabling quantitative detection of HCD. This method features simple operation and low cost, yet it is a non-specific detection technique capable of detecting all dsDNA with low sensitivity. Hence, it is not recommended for standalone use in final product release testing.

2.2 Milestone: qPCR Becomes the Standard Method

After 2000, Real-time Quantitative Polymerase Chain Reaction (qPCR) technology matured. With specific primers and probes, it enables exponential amplification and real-time monitoring of trace DNA, achieving a leap from qualitative analysis to accurate quantification. Featuring ultra-high sensitivity (down to the femtogram (fg) level), strong specificity, a wide linear range and high throughput, qPCR quickly became the industry standard. Since 2015, the United States Pharmacopeia (USP) and the Chinese Pharmacopoeia have all recommended qPCR as the standard method. Its stability, repeatability and regulatory compliance have been verified through long-term practical application.

2.3 Emerging Technology: dPCR for Absolute Quantification (Under Evaluation)

Digital PCR (dPCR) realizes absolute nucleic acid quantification via droplet partitioning combined with Poisson statistics, requiring no standard curve. It also has excellent anti-interference performance against samples and is mostly applied to the determination of gene copy numbers. However, dPCR still faces multiple unresolved challenges when used for routine HCD residual quality control testing:

• High investment cost: dPCR instruments are much more expensive than conventional qPCR devices. In addition, its consumables and reagents have short shelf lives, leading to high overall operating costs.

• Inconsistent test results: Different dPCR platforms generate discrepant data. The technique is highly sensitive to sample pretreatment and pipetting errors, and there is no universal reference material, resulting in inconsistent detection values, especially for samples with low HCD residual levels.

• Increased operational complexity: Extra steps such as droplet generation and chip encapsulation are required, imposing stricter requirements on operators’ standard operating proficiency and laboratory environment. Its linear detection range only covers 3 to 4 orders of magnitude (while qPCR covers 5 to 6 orders of magnitude), so high-concentration samples need multiple dilutions before testing.

dPCR calculates the copy number of target molecules in samples directly through binary counting of positive and negative results based on the Poisson distribution model. This calculation process involves multiple uncertainty factors, including calibration errors of the Poisson distribution, deviations in conversion efficiency, uneven droplet generation, and varying sensitivity to different inhibitors. To apply dPCR to HCD residual quality control of biological products, more efforts are needed to unify test results across different platforms and standardize operational procedures.

03 From Total Quantity Control to Fragment Analysis: Concept Upgrade and Complementary Methods

In recent years, regulatory authorities including the FDA, WHO and the Center for Drug Evaluation (CDE) of China have required not only compliance with total residual HCD limits, but also evaluation of the correlation between DNA fragment size and potential risks. For continuous cell lines such as Vero and MDCK cells, which are non-tumorigenic but possess certain transformation potential, large DNA fragments (>200 bp) are more likely to carry complete oncogene or viral gene sequences, serving as a major source of potential tumorigenic and infective risks. Therefore, the latest detection technologies no longer merely report total HCD values, but conduct comprehensive risk assessment combined with DNA fragment distribution.

In 2006, a consensus was reached at the WHO Expert Group Meeting: limiting the size of residual DNA fragments to below 200 bp can reduce the probability of intact gene sequences existing in residues, thereby further lowering tumorigenic and infective risks. This was the first time worldwide that DNA fragment size was defined as an explicit risk control indicator.

In 2010, the FDA clarified in its guidelines for biological products that risks can be reduced by lowering residual DNA content and shortening DNA fragment size, and required manufacturers to test both DNA content and fragment size distribution in products.

In 2022, the CDE issued the Technical Guidelines for Pharmaceutical Research and Evaluation of In Vivo Gene Therapy Products (Trial), which explicitly mandated the control of both residual DNA content and fragment size, and recommended restricting residual DNA fragments to less than 200 bp. This marked the first official requirement for this indicator in domestic regulatory guidelines in China.

3.1 Capillary Electrophoresis (CE)

Capillary Electrophoresis (CE) is currently one of the mainstream technologies for detecting the size and distribution of host cell residual DNA fragments. Based on the electrophoresis principle, DNA fragments migrate at different rates in an electric field according to their molecular size inside a capillary tube. Intuitive electropherograms are generated after testing: the position of each peak corresponds to a specific fragment size, and the peak height reflects the relative content of the corresponding fragment. CE is especially suitable for process research and optimization, as it can directly evaluate the impact of different process conditions on DNA fragment size.

3.2 Multiplex qPCR Fragment Analysis Technology

Conventional HCD qPCR assays can also indirectly deduce the distribution of DNA fragment sizes.

The HCD fragmentation analysis detection kit developed by Huzhou Shenke Biotechnology Co., Ltd. (abbreviated as HZSKBIOⓇ) is based on the multi-target qPCR principle. Three to five amplicons of different lengths are designed on the host cell genome as "molecular rulers", to quantitatively detect residual DNA fragments of corresponding sizes in samples. For instance, the CHO residual DNA fragment analysis kit is designed with four amplicons of 95 bp, 110 bp, 215 bp and 523 bp; the Vero residual DNA fragment analysis kit adopts four amplicons of 85 bp, 134 bp, 229 bp and 552 bp.

This kit applies the relative abundance ratio analysis method: the quantitative value of the shortest amplicon (84 bp) is set as the reference value (1), and the relative ratio of other amplicons of different lengths is calculated to infer the fragment distribution of residual DNA in samples. The judgment criteria are as follows:

• If the ratios of all amplicons of different lengths are close to 1, it indicates that DNA fragments of various sizes are evenly distributed in the sample, with a large number of intact long-stranded DNA and a low fragmentation degree.

• If the ratios of long-fragment amplicons are significantly lower than those of short-fragment amplicons, it means DNA has been fully degraded, and residues are dominated by small fragments with relatively low risks.

• If only short fragments are detected while no signal is observed for long fragments, the DNA is judged to be highly fragmented.

• If all amplicons show extremely low signals or no detection signal, it proves that the total residual DNA level has been effectively controlled.

  
  

By observing the decreasing trend of ratios from short fragments to long fragments, analysts can intuitively evaluate the fragmentation degree and integrity of residual DNA. This innovative solution is built on the mature qPCR platform, leveraging its stable quantitative capability to realize in-depth risk assessment. Combined with CE, the two technologies complement each other to complete the characterization of DNA fragment sizes.

04 Future Outlook: Standardization and Exploration of New Technologies

4.1 Standardization and Automation

Against the backdrop of increasingly stringent regulatory requirements and diverse types of biotech drugs, HCD detection technologies are evolving toward higher accuracy, efficiency and throughput. The qPCR method has achieved a high degree of standardization, and the widespread application of automated instruments has further improved detection efficiency.

4.2 NGS: Decoding Sequence Information of Residual DNA

Given the complexity of residual HCD in biological products, Next-Generation Sequencing (NGS) is adopted to explore in-depth questions: What are the specific sequences of these residual DNA?Which genes do they carry? Do they contain known oncogenic elements or viral integration sites?

At present, NGS is mainly applied for in-depth characterization and risk identification in the research and development stage. Its large-scale application in routine product release testing faces multiple obstacles, including high costs, complicated data analysis, stringent requirements for operation and bioinformatics analysis, as well as the lack of unified industrial standards and validation guidelines. Nevertheless, NGS is evolving rapidly from a pure research tool to a compliant testing platform. As pharmacopeia organizations including the USP and European Pharmacopoeia (EP) promote the standardization of NGS, and the ICH Q5A guideline recognizes the application of NGS in viral detection, NGS is expected to become an important supplementary tool for HCD risk assessment in the future.

References

[1] Danhua Zhao, Weiying Zong, Wanxin Wu, Yuhua Li, Zongsong Wu, Zhixing Yang, Shouchun Cao. Development and validation of a qPCR assay for the detection of residual host cell DNA in rabies vaccines produced in Vero cells[J]. Frontiers in Bioengineering and Biotechnology, 2025, 13.

[2] Qiu Ran, Zhang Qingmei, Le Yang, Ji Deming, Wu Dongping, Zhang Xiaoyu, Li Fang, Wu Wenyi, Yuan Xiaoling, Zhang Qimeng, Liu Bo, Zhang Zhegang. Establishment, validation and application of a method for analyzing host residual DNA fragment size of pandemic influenza whole-virus inactivated vaccine (H5N1)[J]. Chinese Journal of Biologicals, 2025, 38(12):1451-1468.

[3] Wang Guangyu, Yang Jingqing, Wei Changlong, Zhou Zexin, Yang Zhixing, Huang Yu, Zhou Yong. Preparation of the first national standard for determination of HEK293 cellular DNA content[J]. Chinese Journal of Biologicals, 2024, 37(3):316-321.

[4] Dou Minhua, Zhu Bingmei, Zong Weiying, Wu Xiaoshuang, Yang Zhixing. Study on digital PCR method for establishment of quality control detection method and quantitative reference materials for CAR-T cell therapy products[J]. Chinese Journal of New Drugs, 2021, 30(24):2306-2314.

[5] Wu Gang, Fu Zhihao, Yang Zhixing, Cui Yongfei, Zhang Rongjian, Zong Weiying, Wang Lan. Methodological validation of Taqman probe-based qPCR for detection of NS0 host cell residual DNA[J]. Chinese Pharmaceutical Journal, 2019, 54(24):2001-2009.

[6] Lv Ping, Yang Zhixing, Zhang Hui, Zong Weiying, Wu Wanxin, Wang Tao, Liang Chenggang. Methodological validation of CHO host cell residual DNA detection kit (PCR-Taqman probe method)[J]. Chinese Journal of New Drugs, 2018, 27(21):2519-2526.