Vertical level of blood cell division cycle 42 predicts response and survival benefits to PD-1 inhibitor-based regimen in metastatic colorectal cancer patients

Abstract Cell division cycle 42 (CDC42) regulates the development of colorectal cancer (CRC) by modulating cancer malignant behaviors and facilitating immune escape. Hence, this study aimed to explore the correlation of blood CDC42 with treatment response and survival benefit to programmed cell death-1 (PD-1) inhibitor-based regimens in inoperable metastatic CRC (mCRC) patients. Fifty-seven inoperable mCRC patients who received PD-1 inhibitor-based regimens were recruited. The CDC42 in peripheral blood mononuclear cell (PBMC) was detected using RT-qPCR in inoperable mCRC patients at baseline and after 2-cycle treatment. Besides, PBMC CDC42 in 20 healthy controls (HCs) was also detected. CDC42 was higher in inoperable mCRC patients compared to HCs (p < 0.001). Elevated CDC42 was related to a higher performance status score (p = 0.034), multiple metastatic sites (p = 0.028), and the presence of liver metastasis (p = 0.035) in inoperable mCRC patients. During the 2-cycle treatment, CDC42 was reduced (p < 0.001). Higher CDC42 at baseline (p = 0.016) and after 2-cycle treatment (p = 0.002) were both linked with decreased objective response rate. CDC42 high at baseline was related to shorter progression-free survival (PFS) (p = 0.015) and overall survival (OS) (p = 0.050). Moreover, CDC42 high after 2-cycle treatment was also related to unfavorable PFS (p < 0.001) and OS (p = 0.001). After adjustment using multivariate Cox’s analyses, CDC42 high after 2-cycle treatment independently related to shorter PFS (hazard ratio (HR): 4.129, p < 0.001), and CDC42 reduction ≤230% also independently correlated with shorter OS (HR: 4.038, p < 0.001). The longitudinal change of blood CDC42 during PD-1 inhibitor-based regimen estimates treatment response and survival in inoperable mCRC patients.


Introduction
Metastatic colorectal cancer (mCRC) represents a tremendous medical burden in the clinical aspect, accounting for 20% of newly diagnosed colorectal cancer (CRC) cases [1,2]. Due to the occurrence of distal metastasis, most mCRC patients are inoperable and their management options are limited [3]. Programmed cell death-1 (PD-1) inhibitor has been applied in treating various types of solid tumors (such as lung cancer, renal cancer, bladder cancer, etc.) [4]. Inspiringly, recent clinical trials display that PD-1 inhibitorbased regimen also achieves promising tumor response and survival profiles in inoperable mCRC patients [5][6][7]. However, the objective response rate (ORR) in mCRC patients treated with PD-1 inhibitor-based regimens varies from 31.3% to 43.8%, leaving an unresponsiveness rate above 50% and further impairing their prognosis [8,9]. Hence, it is necessary to identify some potential biomarkers to stratify the inoperable mCRC patients who are beneficial from the PD-1 inhibitor-based regimen before initial therapy and further improve the overall management of inoperable mCRC patients.
Cell division cycle 42 (CDC42) is a small GTPase belonging to the Rho family [10]. Besides, CDC42 plays a critical role in the progression of several cancers by modulating cancer malignant behaviors and facilitating immune escape [11][12][13][14][15]. For instance, CDC42 promotes CRC cell proliferation, migration, and invasion further enhancing the progression of CRC [11][12][13]. Moreover, CDC42 is implicated in the immune escape of malignancies through the suppression of immune response mediated by cytotoxic T cells and natural killer (NK) cells [14,15]. Considering that CDC42 regulates tumor immune escape, and it also directly promotes CRC malignant behaviors, it is hypothesized that CDC42 may exhibit certain values in estimating PD-1 inhibitorbased regimens in inoperable mCRC patients. However, the relevant study is limited.
Therefore, this study detected the blood CDC42 expression in inoperable mCRC patients treated with PD-1 inhibitor-based regimen and aimed to observe the longitudinal change of CDC42 and its correlation with treatment response and survival benefit to PD-1 inhibitor-based regimen in inoperable mCRC patients.

Participants
From January 2019 to December 2021, 57 patients with inoperable mCRC who received PD-1 inhibitor-based regimen at our hospital were recruited. The inclusion criteria were: (1) diagnosed as inoperable mCRC; (2) aged 18 years or older; (3) at least one measurable lesion; (4) received PD-1 inhibitor with or without other treatments. The exclusion criteria were: (1) Eastern Cooperative Oncology Group performance status (ECOG PS) >2; (2) uncontrolled hypertension; (3) had other primary malignancies; (4) complicated with hematological malignancies; (5) complicated with autoimmune diseases; (6) pregnant or breastfeeding women. Besides, 20 healthy subjects were recruited as health controls (HCs). The Ethics Committee permitted the study. The written informed consent was gathered from all participants.

Data and sample collection
The demographics and disease features of inoperable mCRC patients were gathered through case report forms. The peripheral blood (PB) samples of patients with inoperable mCRC were collected at baseline (before treatment) and after 2-cycle treatment (6 weeks). The PB samples of HCs were collected after admission.
The PD-1 inhibitor used was the camrelizumab or sintilimab. The common usage of camrelizumab was 200 mg intravenously at day 1 every 3 weeks. The common usage of sintilimab was 200 mg intravenously at day 1 every 3 weeks. The chemotherapy used was capecitabine and oxaliplatin (CapeOX) or 5-fluorouracil, leucovorin, and oxaliplatin (mFOLFOX6) as described previously [16,17]. The common usage of bevacizumab was 7.5 mg/kg intravenously, once every 3 weeks when combined with CapeOX, once every 2 weeks when combined with FOLFOX6. The common usage of apatinib was oral 370 or 500 mg daily. The dose of apatinib could be reduced to 250 mg per day.
Reverse transcription quantitative polymerase chain reaction (RT-qPCR) CDC42 expression in peripheral blood mononuclear cells (PBMC) was detected by RT-qPCR assay. Briefly, the total RNA extraction was performed by Trizol reagent (Beyotime, China). After extraction, the reverse transcription was completed by applying BeyoRT TM III M-MuLV Reverse Transcriptase (Beyotime, China). Then, the qPCR reaction was conducted by BeyoFast TM SYBR Green qPCR Mix (Beyotime, China). The primers were designed according to a previous study [18]. The thermocycling condition was 1 cycle of 95 C for 60 sec, 40 cycles of 95 C for 15 sec, and 60 C for 60 sec. Moreover, the relative expression of CDC42 was calculated based on the 2 ÀDDCt method using GAPDH as an internal reference [19].

Efficacy
Patients had computerized tomography (CT) or magnetic resonance imaging (MRI) every 4-6 weeks for the first three months and thereafter every 2 months. The treatment response was evaluated at 3 months via the Response Evaluation Criteria in Solid Tumors (RECIST) version 1.1 [20]. The last follow-up date was May 31, 2022. The median follow-up duration was 12.4 months, and the range was 2.5-24.7 months. The progression-free survival (PFS) was classified as the interval from treatment beginning to progression or death. The overall survival (OS) was classified as the interval from treatment beginning to death.

Statistics
Statistical analyses and graph plotting were performed by SPSS v25.1 (IBM Corp., USA) and GraphPad Prism v6.01 (GraphPad Software Inc., USA). CDC42 expression in patients with inoperable mCRC was classified as low or high by the median value. The CDC42 decline was defined as CDC42 at baseline minus CDC42 after a 2-cycle treatment. The unpaired tests were evaluated through the Wilcoxon rank-sum test and Kruskal-Wallis test by ranks. The paired tests were performed via the Wilcoxon Signed Rank Test. The survival data were estimated via Kaplan-Meier curves and log-rank tests. The factors related to PFS and OS were evaluated via univariable and multivariate Cox's regression model with step-forward methods (all factors included). P values of 0Á05 or less showed statistical significance.

Association of PBMC CDC42 expression with clinical features in inoperable mCRC patients
Elevated CDC42 expression was related to a higher ECOG PS score (p ¼ 0.034), multiple metastatic sites (p ¼ 0.028), and the presence of liver metastasis (p ¼ 0.035) in inoperable mCRC patients (Table 2). However, no correlation of CDC42 expression with other clinical features was observed (including age, gender, location, tumor differentiation, lung metastasis, peritoneum metastasis, other metastasis, KRAS mutation, treatment lines, and specific treatment regimens) (all p > 0.05). Moreover, CDC42 expression did not relate to microsatellite instability (MSI)/DNA mismatch repair Figure 1).
Longitudinal change of PBMC CDC42 expression and its correlation with treatment response in inoperable mCRC patients CDC42 expression was reduced during 2-cycle treatment in inoperable mCRC patients (p < 0.001) (Figure 2(A)). Moreover, both higher PBMC CDC42 expression at baseline (p ¼ 0.017) and after 2-cycle treatment (p < 0.001) were related to poorer response (Figure 2(B)). Besides, elevated CDC42 expression at baseline (p ¼ 0.016) and after 2-cycle treatment (p ¼ 0.002) were both linked with the absence of ORR (Figure 2(C)). Similarly, increased CDC42 expression at baseline (p ¼ 0.030) and after 2-cycle treatment (p ¼ 0.001) were correlated with the absence of DCR as well (Figure 2(D)).

Discussion
Until now, one previous study detects the CDC42 expression from tumor tissue and non-tumor tissue in resectable CRC patients by immunohistochemistry (IHC) assay, and it is observed an upregulation of CDC42 expression in tumor tissue compared to paired adjacent non-tumor tissue [21]. Moreover, the elevated tumor CDC42 expression is related to the occurrence of lymph node metastasis and a higher pathological tumor-node-metastasis (TNM) stage in resectable CRC patients [21]. However, the correlation of blood CDC42 expression with disease risk and features in inoperable mCRC is rarely reported. Hence, in the present study, it was observed that PBMC CDC42 expression was upregulated in inoperable mCRC patients compared to HCs. Moreover, increased PBMC CDC42 expression was related to multiple metastases and poorer physical performance status in inoperable mCRC patients. The possible reasons to explain these findings were: (a) CDC42 mediated multiple oncogenic pathways (such as rat sarcoma (Ras)/extracellular regulated protein kinase (ERK) pathway, protein kinase B (AKT) pathway, nuclear factor (NF)-jB pathway, etc.) to promote CRC cell invasion and migration, which further induced the distal metastasis in CRC patients; thus, elevated CDC42 was linked with multiple metastases in inoperable mCRC patients [22][23][24]. (b) CDC42 was related to multiple metastases, which further caused the impairment of multiple systems/organs (such as lung, liver, etc.), and led to higher ECOG PS scores in inoperable mCRC patients.
PD-1 inhibitor, as one immunotherapy targeting immune checkpoints, has been applied in treating inoperable mCRC patients with a combination of systematic therapy and/or targeted therapy [7,[25][26][27]. However, some parts of inoperable mCRC patients report a low response rate. Currently, several biomarkers have been identified to estimate the treatment response of PD-1 inhibitor-based regimens in inoperable mCRC patients [28][29][30]. However, the longitudinal change of blood CDC42 expression and its relationship with response to PD-1 inhibitor-based regimen is rarely evaluated. Therefore, the current study detected the  PBMC CDC42 expression at baseline and after 2-cycle treatment of PD-1 inhibitor-based regimen in inoperable mCRC patients. Then, it was observed that PBMC CDC42 expression was decreased during treatment, also its higher level was related to a poorer response to PD-1 inhibitor-based regimen in inoperable mCRC patients, which could be explained as that elevated CDC42 might promote the occurrence of immune escape through the regulation of T cells and NK cells, which further led to tumor growth and unresponsiveness to PD-1 inhibitor-based regimen in inoperable mCRC patients [14,15,31]. Although mCRC patients in the current study received other treatment regimens aside from PD-1 inhibitors (anti-angiogenesis therapy and/or chemotherapy), CDC42 expression did not differ among mCRC patients with different PD-1 inhibitor-based regimens from our results, indicating the prognostic value of CDC42 in estimating response to PD-1 inhibitors mCRC patients.
In terms of the relationship between CDC42 expression and survival profile, one study reports that increased tumor CDC42 expression is related to shorter disease-free survival (DFS) and OS in resectable CRC patients, but the evidence of blood CDC42 expression in estimating survival profile in inoperable mCRC patients is still lacking [21]. In the current study, the correlation of PBMC CDC42 expression at baseline, CDC42 expression after 2-cycle treatment, and CDC42 reduction during treatment with survival profile were determined. Then, it was discovered that CDC42 high expression at baseline and CDC42 high expression after 2cycle treatment were both related to shorter PFS and OS. Interestingly, CDC42 high expression after 2-cycle treatment seemed to exhibit a more potent prognostic value in estimating survival profile in inoperable mCRC patients than its baseline level. Moreover, it was also observed that CDC42 after 2-cycle treatment (high vs. low) was independently related to shorter PFS in inoperable mCRC patients. These data suggested that monitoring of CDC42 expression during PD-1 inhibitor-based regimen in inoperable mCRC patients may be one way to improve the overall inoperable mCRC management.
However, several limitations still existed in the current study. For instance, the sample size of the present study was limited, thus further study with a larger sample size was necessary. Moreover, further study could enroll non-metastatic CRC patients to further comprehensively analyze CDC42 levels among inoperable mCRC patients, non-metastatic CRC patients, and HCs. Furthermore, the current study only recruited inoperable mCRC patients, while the correlation of CDC42 with PD-1 inhibitor-based regimen's efficacy as post-operation adjuvant therapy in operable CRC patients could be determined in future studies.  In conclusion, blood CDC42 expression is reduced during treatment, and its higher level may relate to unresponsiveness and unsatisfied survival to PD-1 inhibitor-based regimen in inoperable mCRC patients.