Implementation

Overview

Nowadays, cancer, as a complex disease, continues to threaten human health and life with its high incidence and lethality. Among the many types of cancer, non-small cell lung cancer and ovarian cancer are of particular interest due to their unique pathologic features and severe treatment challenges.

Non-small cell lung cancer, as a major component of lung cancer, accounts for about 80% to 85% of all lung cancer cases. Its slow growth but strong spreading power makes the early symptoms not obvious, which often leads to middle to late stage when diagnosed, posing a great threat to patients' lives. Ovarian cancer, as a common malignant tumor of female reproductive system, is second to cervical cancer and uterine cancer in terms of incidence rate, but the mortality rate is the highest among gynecological tumors. Due to the insidious early symptoms of ovarian cancer, most patients have already developed extensive metastasis by the time of diagnosis, which makes the difficulty of treatment increase dramatically. Traditional diagnostic methods for these two cancers are often limited by high misdiagnosis rates, long testing cycles and high costs, resulting in many patients missing out on optimal treatment. To address these challenges, our team is committed to developing a cancer early detection platform based on cancer biomarkers and single molecule detection tools. Our project aims to build a complete cancer early diagnosis system, which includes the whole process from sample processing, probe design, single-molecule detection to data analysis. The target microRNA are screened by machine learning, which in turn leads to the design of relevant probes and the construction of a two-stage amplification system.By converting cancer biomarkers into fluorescent signals for single-molecule fluorescence detection through a two-stage amplification system, we have achieved efficient and accurate detection of microRNA for biomarkers of ovarian cancer and non-small-cell lung cancer, thus providing doctors with a more accurate basis for diagnosis.

Proposed End Users

The target users of this project include the following categories:

Clinicians: As the most direct applicators, clinicians can use our cancer early diagnosis platform to conduct preliminary diagnostic screening of patients to determine whether further in-depth examination is required. The platform can provide intuitive and accurate test results to help doctors make quick decisions and improve diagnostic efficiency.
Patients: For patients suspected of having cancer, the platform can provide earlier and more accurate diagnostic results, thus enabling patients to receive timely treatment and improving the cure rate and quality of life. Meanwhile, for recovered cancer patients, the platform can also be used to monitor the recurrence of the disease and provide early warning of potential risks.
Scientific research: The platform can also be used as a scientific research tool to verify the relationship between a particular cancer and the expression of specific microRNA,helping researchers to explore the molecular mechanism of cancer development.

Envision

As a pioneer in the field of precision diagnostics, our vision is to build a universal platform capable of early screening and accurate diagnosis of various types of cancer. Through continuous technological innovation and optimization, we hope to apply this platform in clinical practice and bring benefits to our patients. Specifically, we hope to achieve the following goals:

Improve diagnostic accuracy: By optimizing the probe design and amplification system, combined with machine learning algorithms, it improves the sensitivity and specificity of detection and reduces the phenomenon of misdiagnosis and missed diagnosis.
Shorten the testing cycle: Optimize the experimental process and improve the efficiency of testing, so that patients can obtain diagnostic results in a shorter period of time.
Reducing the cost of testing: Through technological innovation and large-scale production, the cost of testing will be reduced so that more patients can afford it.
Realize non-invasive testing: Patients do not need to go through the cumbersome and potentially uncomfortable traditional testing methods, patients can be screened for ovarian cancer, non-small cell lung cancer and other cancers at an early stage through simple blood collection,which improves the testing experience and comfort of patients.
Promote scientific research: Use the platform as a scientific research tool to promote the in-depth development of cancer biology research, accelerate the discovery and validation process of new cancer markers, and provide a solid theoretical foundation for the precise treatment of cancer.

Implement

The implementation of this project consists of the following key steps:

Probe design and synthesis
First, we utilize machine learning algorithms to screen out microRNA that are highly correlated with specific cancers as targets from massive data. Then, we design pluggable probes according to the sequence characteristics of the targets.
Amplification system design and optimization
In order to realize highly sensitive detection of cancer biomarkers, we designed a self-enhancement system based on strand substitution reaction and a self-enhancement system based on CRISPR system. The two-stage amplification system converts cancer biomarkers into fluorescent signals to meet the needs of single-molecule detection.
Single-molecule assay process
We have established a complete single-molecule assay procedure, including sample processing, probe addition, single-molecule fluorescence microscopy and other steps. During the experimental process, we focus on controlling the experimental conditions and environmental factors to ensure the stability and reproducibility of the test results. Meanwhile, we have also established a strict data analysis and quality control process to ensure the accuracy and reliability of the assay results.
Synthetic target validation and real target detection
After the completion of probe synthesis, we utilize synthetic targets for initial validation to ensure the functionality and specificity of the probes. Then, we turn to the detection of real targets, i.e., microRNA extraction and detection using known cancer cell samples, tissue samples of cancer patients and blood samples to ensure the accuracy and reliability of the detection results.
Clinical validation and popularization
After completing the experimental validation, we will actively cooperate with hospitals to conduct clinical trials. By collecting and analyzing the test data from a large number of clinical samples, we will continue to optimize the performance of the platform and improve its clinical application value. At the same time, we will also strengthen cooperation and communication with other medical institutions and research institutes, so that it can benefit more patients and researchers.

Safety Considerations

In the course of the project, we attach great importance to safety and have taken a number of measures to ensure the safety of experimental and clinical applications:

Safety of biological samples: In the process of sample handling and testing, we strictly follow the biosafety operation procedures to ensure that the samples will not cause contamination to the environment and personnel. Meanwhile, we have also established a strict sample management system and traceability system to ensure the safety and traceability of samples.
Instrument and equipment safety: The use and maintenance of high-precision instruments and equipment, such as single-molecule fluorescence microscopes, require strict compliance with operating procedures and regular calibration and maintenance to ensure their normal operation and reduce the risk of failure. In addition, we have established an emergency response mechanism for instrumentation to cope with possible emergencies.
Experimental process safety: We pay special attention to the safety of the entire experimental process from the synthesized target to the actual biological samples (including cancer cells, patient tissues and blood), and ensure that each step of the operation follows the strict safety protocols to reduce the safety risks during the experimental process.
Data privacy and security: In the process of data collection, processing and analysis, we strictly comply with relevant laws and regulations and ethical norms to ensure the privacy of patients and the security of medical data. During project implementation, we continuously comply with ethical norms and legal and regulatory requirements to ensure compliance and ethics of the research.

Challenges

Despite the broad application prospects and clinical importance of our project, we still face many challenges in the implementation process. The following are the major challenges that we need to seriously consider and work to overcome:

Technical challenges
False positives and background noise: The problem of false positives mainly stems from accidental binding of probes. Although we have screened highly specific microRNA targets through machine learning, it is still possible for probes to bind non-specifically to non-target molecules in complex biological environments, leading to the generation of false-positive signals. To overcome this challenge, we need to further optimize the probe design to enhance its binding specificity to target molecules and develop more stringent validation processes to distinguish true signals from false-positive signals.
Background noise is another issue that cannot be ignored. During the detection process of single-molecule fluorescence microscopy, background noise may originate from a variety of factors, such as the noise of the instrument itself, contamination during sample handling, and environmental factors. These noises can interfere with the recognition of effective fluorescent spots and reduce the sensitivity of detection. In order to reduce the influence of background noise, we need to fine-tune the control of the experimental process, and at the same time use advanced signal processing techniques to enhance the effective signal and suppress the background noise.

Example Data Challenges
Our diagnostic platform currently relies heavily on tissue sample databases for screening and analysis of cancer biomarkers. However, the source of tissue samples and the diversity of sampling sites have a significant impact on the expression of molecular markers, which may lead to some limitations in the specificity of our test results.
Another key challenge lies in the uneven distribution of samples in our database. Specifically, the number of diseased samples is much higher than that of non-diseased or healthy controls. This imbalance not only affects our accuracy and reliability in screening targets and performing machine learning, but may also lead to biased models that are overly inclined to predict disease status.
To address the problem of database imbalance, we chose to use a balancing function as a stopgap measure. This strategy has the advantage of improving the accuracy and performance of the model to a large extent, especially in identifying non-disease samples, but is not very rigorous. In future studies, we will continue to explore more effective data balancing methods to further improve the accuracy of our study.
High requirements of operation technologys
From sample collection, processing to detection and analysis, every step of the operation requires a high degree of professional skills and strict quality control. The complexity of single-molecule fluorescence microscopy in vitro assay technology puts higher requirements on the technical level of operators. Therefore, we need to strengthen the training and technical support for our team members to ensure the accuracy and reproducibility of the test results.

In summary, although our project has great potential and application value in the field of early cancer diagnosis, we still face many challenges in the implementation process. We will meet these challenges with scientific rigor and the spirit of continuous innovation, and strive to promote the smooth implementation and wide application of the project.