A Research Protocol for Evaluating the Efficacy of Perfused Chemotherapeutic Agents for Bladder Cancer Based on Organoid Technology

1. Background Bladder cancer is a malignant tumor originating from the uroepithelium of the bladder, accounting for the first place in the incidence of genitourinary tumors in China. Among them, bladder uroepithelial cancer is the most common, accounting for more than 90% of bladder cancers. According to the depth of tumor invasion into the bladder and the prognostic characteristics, bladder cancer is clinically classified into non-muscle-invasive bladder cancer (NMIBC) and muscle-invasive bladder cancer (MIBC), with NMIBC as the main type, accounting for about 75% of all bladder cancer patients . NMIBC is usually treated with Transurethral Resection of Bladder Tumor (TURBt). Despite being in the early stages of the disease, NMIBC has a high recurrence rate, with studies reporting a 5-year recurrence rate of approximately 70% . Therefore, postoperative bladder perfusion chemotherapy has become a key therapeutic measure to prevent recurrence of bladder cancer.For more than 60 years, people have been exploring more effective bladder perfusion drugs and methods.In 1961, Thiotepa was used for bladder perfusion therapy, and gained a certain degree of efficacy in preventing the recurrence of superficial tumors of the bladder.In 1976, Morales believed that bladder cancer was related to immune deficiencies, and accordingly BCG bladder perfusion therapy has been extremely successful, creating a new way of immunotherapy for bladder tumors, which has been widely valued and is still considered the most effective bladder perfusion drug. However, BCG instillation can cause serious side effects, including urethral and prostatic granulomas, bladder irritation, fever, hematuria and other local manifestations and systemic influenza-like symptoms, resulting in intolerance and interruption of treatment in some patients, which has limited the wide clinical application of BCG. Studies have shown that intravesical instillation of chemotherapeutic agents in the bladder can reduce the near-term recurrence rate of superficial bladder cancer by about 15%-20% and the long-term recurrence rate by about 6% . Currently there are many choices of chemotherapeutic agents in bladder instillation, including (1) pirenzolubicin; (2) gemcitabine; (3) mitomycin; (4) epirubicin; (5) doxorubicin; (6) alternating bladder-infusion chemotherapy with gemcitabine and pirenzolubicin; and (7) other combined-infusion chemotherapies. However, due to the wide variety of chemotherapeutic drugs, drug selection mostly relies on clinical experience or certain research results, there is no uniform standard, and it is blind, empirical, and randomized. Chinese urological disease diagnosis and treatment guidelines also did not provide the selection of perfusion drugs, and individual patients for some chemotherapeutic drugs show natural resistance, when the clinic determines that the patient is not sensitive to the application of the drug, the drug has produced serious toxic side effects, and even lead to the phenomenon of multi-drug resistance (multi-drug resistance, MDR) produced, so that the clinical The recurrence rate is still as high as 36%-44%, and at the same time, the tumor of this group of patients progresses rapidly, losing the opportunity to re-select the treatment method. Therefore, how to avoid the selection of primary drug-resistant drugs and directly choose drugs with high sensitivity to achieve individualized chemotherapy has become a hot spot in the research of bladder perfusion chemotherapy. The realization of precision tumor therapy greatly depends on the detection of drug sensitivity. Through precise individualized chemotherapeutic drug screening experiments, the most effective and least toxic therapeutic regimen can be judged for each patient before the start of treatment, so as to propose a chemotherapeutic regimen for a single patient is a new direction of research to realize the precision treatment of bladder cancer and to improve the efficiency of bladder cancer chemotherapy. Previously, the more commonly used preclinical models are traditional tumor cell lines and human-derived tumor tissue xenografts (PDX). Tumor cell line culture method is simple but insufficient to simulate the growth state of tumor cells in patients, and the drugs screened by its drug screening system have low value for clinical application.PDX, although it can simulate in vivo tumor characteristics and preserve the tumor microenvironment, has obvious limitations such as low stable tumorigenicity, long modeling and evaluation period (half a year to one year), time-consuming and laborious, and it is difficult to generate and use for high-throughput drug screening. Therefore, the development of drug-sensitivity assay models that can mimic the heterogeneity and complexity of bladder cancer has become necessary in order to develop more personalized therapeutic and preventive strategies to minimize risk and optimize the effectiveness of medical interventions by targeting the unique genetic, environmental and lifestyle characteristics of individuals. Patient-Derived Organoids (PDO) are 3D organoid structures formed by stem cells self-assembled in vitro, which can be differentiated into multiple organ-specific cell types and can exhibit cell-cell and cell-surrounding matrix interactions and spatial location patterns, recreating in vitro some of the key functions and structures of real organs, and having a stable phenotype. structures with stable phenotypic and genetic characteristics. Compared with two-dimensional tumor cell lines and PDX, tumor organoids can be cultured directly using the patient's own tissues, and at the same time, these organoids can well replicate some of the key characteristics of the primary tumors, retain the pathomorphology and biological mechanisms of the patient's tissues, and preserve the heterogeneity of the tumor tissues and a more realistic tumor microenvironment, as well as having a short growth cycle. It is helpful for its use in clinical cancer patients for drug sensitivity testing of radiotherapy drugs, molecular targeting drugs, anti-tumor antibodies and other drugs, to predict the patient's responsiveness to drugs, with the potential to assist in clinical treatment decisions. In 2018, Science reported a study on the use of metastatic gastrointestinal tumor-like organs for drug sensitivity testing, which comparatively analyzed the differences in sensitivity between 21 clinical patients and their corresponding PDOs to a series of targeted and chemotherapeutic drugs, and the results showed strong consistency between the two. In comparison with the actual patient outcomes, the PDOs were well predicted (sensitivity 100%, specificity 93%, positive predictive value 88%, and negative predictive value 100%).20 In 2020, Yao Y et al constructed 96 rectal cancer organoids using biopsies from 112 cases of locally advanced rectal cancer, and selected 80 of them to test their response to radiotherapy, and the results showed that rectal cancer organoids were sensitive to radiotherapy and the patient's clinical response. The results showed that the sensitivity of rectal cancer organoids to radiotherapy was highly consistent with the clinical response of patients (sensitivity 78%, specificity 92%, accuracy 84%). Subsequently, in tumors such as gastric cancer and breast cancer, the concordance between PDO and tumor patients' response to drugs was also found. Yan HHN et al constructed a gastric cancer organoid library using tumor tissues, paracancerous tissues and lymph node metastases from 34 gastric cancer patients. Two of them developed tumor metastases and underwent a combination of cisplatin and 5-FU after surgery, both of which responded well. The other case received chemotherapy before surgery and did not respond to capecitabine after surgery. Examination of the sensitivity of the corresponding compounds of PDO in these three cases showed that the drug sensitivity of PDO was in perfect agreement with the clinical response of each patient.Guillen KP et al constructed PDX and PDO using tumor samples from endocrine therapy-resistant, relapsed, and metastatic breast cancer patients, and these samples were examined histomorphologically, genomically, and drug sensitivity. The results showed that both breast cancer PDX and PDO were highly reductive of the histobiological and genomic properties of their source tumors, and both responded consistently to anti-tumor drugs. A patient with triple-negative breast cancer in stage IIA in this study developed liver metastases about 1 year after undergoing preoperative chemotherapy and surgical treatment. The investigators subjected the constructs PDO and PDX to ex vivo drug sensitivity testing and found that the microtubule inhibitor eribulin had the best therapeutic effect. Based on this result, patients were instructed to undergo treatment with eribulin, which resulted in complete remission of liver metastases for nearly 5 months after dosing. These studies confirm to some extent the possibility of PDO to guide the medication of patients with clinical tumors. It has also been shown that by comparing the difference in response to drugs between normal-like organs and PDO, it has been found that drugs with high selectivity help to reduce toxic side effects in clinical patients. Thus, it is clear that drug sensitivity testing by PDO to discover the most appropriate drug regimen will help to improve the clinical efficacy of tumor patients, reduce toxic side effects, risk of drug resistance, and chances of tumor recurrence, and maximize the benefits to patients. Therefore, relying on patient-derived bladder cancer organoid models, we will carry out non-muscle invasive bladder cancer perfusion chemotherapy drug sensitivity testing experiments, establish a standardized drug sensitivity testing system for bladder cancer organoids, and formulate screening standards for drug sensitivity testing of bladder cancer organoids, so as to screen the optimal drug combination regimen, assist in the clinical development of new individualized treatment protocols, and carry out multi-center clinical validation to achieve a true "Substitute drug testing".
2. Research questions and objectives Research questions: This study compares the one-year tumour-free recurrence survival rate and the three-year tumour-free recurrence survival rate of patients in the organoid-sensitive drug infusion group, the organoid-unsensitive drug infusion group, and the BCG infusion group by means of a cohort study, and the difference in the recurrence rate among the three groups is compared by one-way K-M survival analysis. Assessment to evaluate the value of tumour-like organ drug sensitivity assays in guiding individual perfusion chemotherapy for bladder cancer. Research objective: to assess the application value of tumour-like organ drug sensitivity experiments in guiding individual perfusion chemotherapy for bladder cancer by observing the clinical efficacy of tumour-like organ drug sensitivity method in guiding postoperative perfusion chemotherapy for bladder cancer patients. See Outcome Measures for details
3. Research methodology 3.1 Study design This protocol is based on the Technical Guidelines for Clinical Trials of Antineoplastic Drugs, and adopts a multicentre, cohort study.

3.1.1 Grouping See Groups and Interventions for details 3.1.2 Bladder cancer organoid preparation

1. Bladder cancer sample collection The sample collection process is in accordance with hospital ethics and signed informed consent. Patients enrolled in the experiment must be pathologically diagnosed with bladder cancer, and the size of tumour samples taken out surgically should be larger than 0.8 cm*3 as far as possible, for pathological verification, organoid culture, drug sensitivity test, and so on. Tumour tissues should be collected with attention to tissue ischemia time and cryopreservation, and the collection time should be within 5 min when the tissues are removed from the human blood circulation, so as to avoid stress necrosis of the tumour tissues due to early hypoxia.
2. Bladder cancer organ preparation In the process of tissue processing, physical crushing and collagenase digestion were used to process the tissue samples, and the processed and digested organoids needed to be filtered through cell sieves of different diameters before culture, so as to make the obtained organoids uniform in size and facilitate the unification of standards. Then they were planted in three-dimensional matrix gel and cultured in special medium.
3. Expansion and culture of bladder cancer organoids The growth status of the organoids was observed microscopically, and immunohistochemistry and immunofluorescence were used to characterize the organoids, to confirm the presence of tumour cells and to evaluate their morphology. Bladder cancer carcinoid organs were passaged and frozen.

3.1.3 Bladder cancer tumour organoid drug sensitivity testing After observing the number of organoids under microscope to reach the number that can be subjected to drug sensitivity testing, the organoid medium was aspirated and discarded. Add 3 mL of digestive solution to the petri dish, blow the droplets apart, digest at 37℃, blow the organoids and observe under the microscope, when the organoids are digested into a homogeneous cell mass containing 2-3 cells, add 2 times the volume of the digestive solution to terminate the digestion. 1500 rpm, centrifuge the cells for 3 minutes, and then aspirate and discard the supernatant.

Resuspend the cells with medium, try not to produce air bubbles, and dilute the medium to gel 1:1.5 ratio, then inoculate the mixed gel at 10uL/well with a single-channel pipette into pre-cooled 384-well plates with 3,000 organoids per well (this operation was performed on ice). After mixing the medium with the gum at a ratio of 1:1 (40uL:40uL), the medium was dispensed into the 384-well plate at 10uL/well as a blank group. After gently tapping the 384-well plate so that the gel droplets lay flat on the bottom of the wells, the plate was placed in an incubator for 30 min to allow the gel droplets to solidify and pre-warmed BCOs medium was added at 40 uL/well. To the drug test wells as well as to the empty wells around the control wells, 40 uL of PBS solution was added and placed in a carbon dioxide incubator to prevent the liquid from evaporating and causing the gel drops to dry out.

After 3 days of incubation and observation, microscopic observation of the status of the sample in each well was carried out for half-volume fluid exchange.

Design the drug sensitivity test protocol according to the requirements of the following table, set up the control group, blank group and drug test group, the drug was dissolved according to the instructions and diluted 4-fold into 6 concentrations with organoid medium, respectively; using luminescent cell viability assay, the drug was tested at different concentrations of the semi-inhibitory concentration (IC50), inhibition of tumour cells (IR), the degree of deviation of the IC50 from the blood drug concentration, AUC ( area under the dose-response curve), the evaluation criteria of the results of the drug sensitivity assay take IC50 and IR as the main indexes, together with the degree of deviation of IC50 from the blood drug concentration and the AUC results as the auxiliary judgement (for details, please refer to 3.10.3), to assess the effects on the survival and proliferation of the class of organisms, and to give the clinical guidance of the use of the drug.

Table 1 Methods for grouping drug sensitivity tests Grouping method of drug sensitivity testing Blank group: No organoid, no drug added, 6 replicate wells per plate. Control group: Inoculated with organoids, no added drug exposure. Drug test group: Inoculated with organoid, add drug exposure. 3.1.4 Chemotherapy drug combination regimens Table 2 Perfusion chemotherapy drugs Gemcitabine □ Pirorubicin □ Epirubicin □ Silvestrol □ Adriamycin(doxorubicin) □ 3.2 Study site and study population (source and inclusion exclusion criteria) See study design for details 3.3 Study Variables (Factors) and Measurements The positive control group was not subjected to organoid culture as well as organoid drug sensitivity testing.

The remaining patients were subjected to organoid drug sensitivity testing in accordance with the drug sensitivity testing protocol, using luminescent cell viability assay to test the semi-inhibitory concentration (IC50) of the drug, inhibition rate (IR) of the tumour cells, the degree of deviation of the IC50 from the blood drug concentration, and the AUC (the area under the dose-response curve) at different concentrations, and the evaluation criteria of the results of drug sensitivity testing were based on the IC50,IR as the main indexes with the IC50 deviation from blood drug concentration, AUC results as auxiliary judgement (see 3.10.3 for details), to assess its effect on the survival and proliferation of class organs.

Patients who underwent organoid culture and drug sensitivity testing underwent bladder instillation of chemotherapeutic agents according to the experience of clinicians. The five perfused drugs included gemcitabine, piroxicam, epirubicin, mitomycin, and adriamycin (doxorubicin). According to the sensitivity of their perfusion drugs in organoid drug sensitivity test, they were divided into organoid sensitive drug perfusion group and organoid non-sensitive drug perfusion group. Perfusion regimen: including induction perfusion chemotherapy (4-8 weeks after surgery, 1 time per week, 4 times in total) and maintenance perfusion chemotherapy (1 time per month, 11 times in total).

BCG infusion regimen: 6 induction infusions followed by 3 booster infusions once every 2 weeks to maintain a good immune response, followed by 10 maintenance infusions once a month for a total of 19 infusions over 1 year.

Clinical measurements: the two groups of patients in the organoid-sensitive drug perfusion group and the organoid-non-sensitive drug perfusion group were followed up and analysed for the primary and secondary study endpoints, and one-way Kaplan-Meier survival analysis was performed to compare the differences in recurrence rates between the two groups. The COX risk proportion model was used to analyse the recurrence risk ratio of patients' gender, age, tumour stage and grade between the two groups.

3.4 Study outcomes RECIST1.1 criteria were used for efficacy evaluation, and Choi criteria and mRECIST criteria were used as auxiliary efficacy evaluation criteria.

Primary study endpoints: one-year tumour recurrence-free survival rate, three-year tumour recurrence-free survival rate.

Secondary endpoints: one-year tumour progression-free survival rate, three-year tumour progression-free survival rate.

3.5 Follow-up The two groups of patients in the organ-like sensitive drug perfusion group and the organ-like non-sensitive drug perfusion group were followed up and analysed for the primary and secondary study endpoints.

Cystoscopy is the method of choice when reviewing patients with NMIBC and there is no non-invasive alternative to cystoscopy. Therefore, NMIBC follow-up should be based on regular cystoscopy. If suspicious lesions of the bladder mucosa are found during the examination, biopsy should be performed to clarify the pathological findings. Urine exfoliative cytology, CT/CTU or MRI/MRU are performed if necessary, but none of them can completely replace cystoscopy.

The cystoscopy review plan: low-risk NMIBC is done once at 3 months and 12 months after surgery in the 1st year, and once a year thereafter until the 3rd year; intermediate-risk NMIBC is done once at 3 months, 6 months, and 12 months after surgery, every 6 months in the 2nd year, and every year thereafter until the 3rd year; high-risk NMIBC is done every 3 months in the first 2 years, and every 6 months thereafter until the 3rd year.

3.6 Sample size A sample size of 293 cases was sufficient to fulfil the requirements for the study endpoints.

There were 73 cases in the organoid non-sensitive drug perfusion group and 110 cases each in the organoid sensitive drug perfusion group and positive control group.

Calculation basis:

1. Calculation of sample size for the one-year tumour-free survival rate study endpoint: the one-year tumour-free survival rate is expected to be 73% in the organoid-sensitive drug infusion group, and 50% in the organoid non-sensitive drug infusion group, with unilateral α=0.05, β=0.2, and the ratio of the sample sizes of the two groups is 1 (experimental group: control group), the sample sizes of the two groups are 55 cases in the experimental group, 55 cases in the control group, and 110 cases in the positive control group. 55 cases in the control group, totalling 110 cases, and considering 20% shedding, a total sample size of 138 cases is required.
2. Calculate the sample size according to the endpoint of the three-year tumour-free survival rate study: it is expected that the three-year tumour-free survival rate of the organoid-sensitive drug infusion group will be 63%, and the three-year tumour-free survival rate of the organoid-unsensitive drug infusion group will be 40%, and the unilateral α=0.05, β=0.2, and the ratio of the sample sizes of the two groups will be 1 (experimental group: control group), so that the sample sizes of the two groups will be calculated as 58 cases in the experimental group, 58 cases in the control group, and 11 cases in the control group, taking into account 20% shedding, and a total sample size of 138 cases will be required. 58 cases in the control group, totalling 116 cases, and considering 20% shedding, a total sample size of 146 cases was required.
3. According to the results of the literature, the one-year tumour-free survival rate of the positive control group is 75%, and according to the pre-test, the one-year tumour-free survival rate of the organoid-sensitive drug infusion group is expected to be 73%, and the non-inferiority cut-off value is 0.2 (control group-experimental group), and the ratio of the sample sizes of the two groups is one with the unilateral alpha being 0.05 and beta being 0.2, the sample size of the two groups is calculated as 74 cases of the organoid sensitive drug infusion group, 74 cases of the positive control group, and a total of 116 cases of the experimental group. 74 cases, and 74 cases in the positive control group, for a total of 148 cases. Considering 20% shedding, the total sample size required is 186 cases.
4. According to the results of the literature, the three-year tumour-free survival rate of the positive control group is 65%, and according to the pre-test, the three-year tumour-free survival rate of the organoid-sensitive drug infusion group is expected to be 63%, and the non-inferiority cut-off value is 0.2 (control group-test group), and the ratio of the sample sizes of the two groups is 1 if the unilateral alpha is 0.05 and the beta is 0.2, the sample size of the two groups is calculated as 88 cases of the organoid-sensitive drug infusion group and 88 cases of the positive control group. 88 cases and 88 cases in the positive control group, totalling 176 cases, and taking into account 20% shedding, a total sample size of 220 cases was required.

3.7 Flow chart of research techniques This section (pictures) describes the technical route of the project. 3.8 Data collection and management Data were collected by observation and recording, and the original data were checked item by item to ensure that the data were consistent with those in the original records. At the end of the experiment, it was handed over to the statistician for statistical analysis.

3.9 Statistical analysis methods Measured data are generally descriptive statistics with mean, standard deviation, median, minimum and maximum values. For categorical variables, descriptive statistics will be performed according to frequency and percentage.

Statistical software SPSS was applied for statistical analysis. Measurement data were expressed as mean ± standard deviation (x ± s), t-test, one-way ANOVA (post hoc test two-by-two comparisons using the LSD method), chi-square test was used for comparison of rates, and Spearman's method was used for correlation analyses. p < 0.05 was considered that the difference was statistically significant.

3.10 Quality control 3.10.1 Class organ HE staining and immunohistochemistry results Conform to the pathological test results, indicating that the constructed organoid and tumour specimens can better retain pathological stability.

3.10.2 Organoid WES or RNA-Seq analysis Constructed organoid and tumour specimens can better retain genetic stability. 3.10.3 Evaluation criteria for drug sensitivity test results The evaluation standard of drug sensitivity test results has IC50,IR as the main index, with the degree of deviation between IC50 and blood drug concentration, AUC results as an auxiliary judgement, when the IC50, IR sensitive indexes are met at the same time, it is regarded as sensitive; when it is not possible to meet the IC50, IR indexes at the same time, IC50 combined with the IC50 and the need for the concentration of the ratio of the smaller degree of difference in the results of the sensitivity.

3.10.3.1 IC50 evaluation criteria. IC50>20× Not sensitive

1×<IC50≤20× Low sensitivity 0.5×<IC50≤1× Moderately sensitive IC50≤0.5× highly sensitive The sensitive group includes strongly sensitive and moderately sensitive, and the insensitive group includes weakly sensitive and resistant.

3.10.3.2 Evaluation criteria for IR tumour inhibition rate . IR <30% is insensitive and classified as negative; IR≥30% is sensitive and classified as positive, in which 30%≤IR<50% is low sensitivity, 50%≤IR<70% is moderate sensitivity, and IR≥70% is high sensitivity.

Sensitivity rate = number of cases with IR ≥ 30% / total number of cases in the group.

3.10.3.3 Calculation of the degree of deviation of IC50 from blood concentration: the degree of deviation from blood concentration refers to the degree of difference between the IC50 value obtained from the drug sensitivity test and the blood concentration, and is calculated as IC50/blood concentration.

The smaller the ratio, the more effective the drug. 3.10.3.4 Criteria for evaluation of AUC (area under the drug-time curve) results: The specific calculation of AUC usually involves integration of the concentration-time curve.

AUC is closely related to the dose-response relationship. In the dose-response curve of drug efficacy, the AUC is a composite indicator of the overall effect of a drug on an organism between dose and response. A larger AUC is usually associated with a better therapeutic effect.

3.11 Post-study treatment of organoid The organoids constructed successfully in this study will be uniformly destroyed and disposed of after the study in accordance with the norms of biomedical waste disposal.