Impact of Molecular Status on Cytoreductive Surgery for Peritoneal Metastases from Colorectal Cancer

Abstract

Colorectal cancer peritoneal metastases (CRC-PM) are present in 5 to 15% of instances of CRC, and the overall survival (OS) of patients with CRC-PM is much lower than that of patients with other isolated metastatic locations. In recent years, the introduction of cytoreductive surgery (CRS) in conjunction with hyperthermic intraperitoneal chemotherapy has resulted in a significant improvement in CRC-PM patients' OS. Despite this, a significant proportion of CRS patients continue to suffer complications of grades III to V or even die during the perioperative period. Early diagnosis, optimization of patient selection criteria, and refining of individualized combination therapy are necessary for these patients. In this review, we evaluate studies examining the relationship between molecular status and CRS in CRC-PM. Our objective is to gain a comprehensive understanding of how the altered molecular status of CRC-PM impacts CRS, which could increase the likelihood of tailored therapy in the future.

Colorectal cancer (CRC) is the third most common malignancy worldwide (10.0% of the total cases), ¹ with a significant increase in incidence compared with 2018 (6.1%). ² Also, CRC has been the second leading cause of cancer-related deaths for many years (2018: 9.2% of the total cancer deaths; 2020: 9.4%). ^{1 2} CRC incidence and mortality continue to rise rapidly in many developing countries, and the public health burden of CRC is expected to increase by 60%, with more than 2.2 million newly diagnosed cases and 1.1 million deaths by 2030. ³

Approximately 25% of CRC patients have metastases at the time of diagnosis, most commonly to the liver via hematogenous metastases. ⁴ Peritoneal seeding of cancer cells may lead to peritoneal carcinomatosis (PC), which is quite common in patients with CRC, ⁵ and peritoneal metastases (PM) from CRC are found in 5 to 15% of CRC cases. ^{6 7 8} Overall survival (OS) is significantly shorter in patients with CRC-PM than in patients with other isolated metastatic sites. ^{9 10}

Due to the relatively low sensitivity and accuracy of diagnostic techniques and the relative lack of therapeutic options in the past, patients with CRC-PM were considered incurable and ultimately only palliative care was an option. In recent years, the introduction of cytoreductive surgery (CRS) combined with hyperthermic intraperitoneal chemotherapy (HIPEC) has dramatically increased the OS rate of patients with CRC-PM ( Fig. 1 ). ^{11 12 13 14} The cure rates in selected patients (16%) even approached that after colorectal liver metastasectomy. ¹⁵ However, while seeing such encouraging clinical results, it is also important to see that 34% of patients still develop grades III to V complications after CRS, and 3 to 4% die in the perioperative period. ^{16 17} Therefore, CRS for CRC-PM has been the focus of research, and it is essential to clarify which patients can obtain the best outcome.

Fig. 1.

Cytoreductive surgery for patients with peritoneal metastases from colorectal cancer.

Based on the assumption that the clinical behavior of PM in CRC is determined by biological mechanisms, molecular biomarkers read-outs are a very promising adjunct to address these clinical needs. ^{18 19 20} Understanding the impact of molecular status on CRS and identifying molecular biomarkers that are predictive or prognostic of treatment outcomes could further enhance the future possibilities for personalized treatment.

Despite the increasing use and tremendous success of molecular biomarkers as proven tools for patient and treatment selection in the treatment of CRC (e.g., ERBB2, RAS, BRAF, TP53, APC ), ^{21 22 23 24} relatively few studies have explored the role of molecular status in response to CRS in CRC-PM patients. Therefore, this review aims to provide an overview of the altered molecular status of CRC-PM patients treated with CRS and to assess the role of these molecular biomarkers in patient selection, the guidance of combination therapy, patient outcomes, and patient prognosis prediction in CRS treatment.

Molecular Status Associated With Outcome of Cytoreductive Surgery for Colorectal Cancer With Peritoneal Metastases

RAS

The Rat sarcoma viral oncogene homolog (RAS) family belongs to the small GTPases and contains three human genes: V-K _i -ras2 Kirsten rat sarcoma viral oncogene homolog (KRAS), neuroblastoma-RAS (NRAS), and Harvey-RAS (HRAS). ²⁵ In the current treatment protocol for CRC-PM patients, RAS and b-viral oncogene homolog B1 (BRAF) mutation status are the sole molecular indicators used to assess eligibility for cytoreduction or maintenance therapy with epidermal growth factor receptor (EGFR) inhibitors. ^{26 27} KRAS is the predominant mutant subtype in CRC (86%), whereas NRAS mutations are rare (14%), and no HRAS mutations are detected. ²⁸ RAS is one of the earliest mutated genes in many cancers, and the RAS oncogene family has been shown to contribute to the development and progression of metastatic CRC (mCRC), where mutations lead to dysregulation of cell proliferation downstream of the EGFR. ^{29 30} Therefore, exploring the mutation status of RAS in CRC-PM patients may not only guide the selection of patients with CRS but also predict the survival rate and recurrence status of patients, which can be used in combination with the above information for individualized treatment and follow-up strategies. ²⁸

In a multicenter study by Breuer et al, ³¹ including 505 CRC-PM patients undergoing CRS-HIPEC, the presence of KRAS or NRAS mutations [hazard ratio (HR) = 2.30, 95% confidence interval (CI) = 1.58–3.33; p  < 0.001] was associated with shorter disease-free survival (DFS), and the presence of RAS mutations ( KRAS or NRAS ) [odds ratio (OR) = 2.42, 95% CI = 1.11–5.47; p  = 0.03] has been identified as a predictor of peritoneal recurrence after CRS-HIPEC. This finding was also confirmed by another study in which Morgan et al ³² compared 22 RAS mutations with 23 patients with RAS wild-type (WT) CRS-HIPEC and found that RAS mutations were associated with reduced DFS at 5.4 and 12.5 months.

In recent years, the question “Is there a significant correlation between RAS mutation status and OS in CRC-PM patients undergoing CRS-HIPEC?” has been debated. Some studies have suggested that patients with RAS mutations are associated with a shorter OS. In a retrospective study by Schneider et al ³³ on CRC-PM patients undergoing CRS-HIPEC, a reduced OS was found in 186 patients with RAS / RAF mutation compared with 192 patients with RAS / RAF WT. Arjona-Sanchez et al ³⁴ showed that only RAS mutational status (HR = 2.024; p  = 0.045) and peritoneal surface disease severity score (PSDSS) stage (HR = 2.90; p  = 0.009) were shown to be independent factors for OS. Early PSDSS stages I and II associated to RAS mutations impaired their OS with no significant differences with PSDSS stage III OS ( p   >  0.05).

Another part of the article argues that RAS mutation status alone should not be used to select CRS patients. Recently reported 152 CRC-PM patients selected to undergo perioperative systemic chemotherapy and CRS with or without HIPEC, Baratti et al ³⁵ demonstrated that WT KRAS / NRAS / BRAF patients showed a median survival of 49.7 months for patients with WT KRAS / NRAS / BRAF and 49.3 months for patients with KRAS or NRAS mutations and WT BRAF ( p  = 0.71). Morgan et al ³² reported no significant difference in OS between the RAS -mutant and RAS -WT CRS-HIPEC groups. In another study that included 195 patients, Bhullar et al ³⁶ showed that RAS ( p  = 0.21) mutation status did not predict OS for patients receiving CRS for CRC-PM.

KRAS

As KRAS mutations are the most prevalent in the RAS family, some studies have only examined the impact of KRAS in CRS patients with CRC-PM. We investigated some relevant studies to investigate the question “Can a single KRAS mutation status predict OS in patients with CRC-PM treated with CRS?” Tonello et al ³⁷ recruited 437 CRC-PM patients treated with CRS from 13 Italian centers, 46.2% of whom had KRAS mutations, and KRAS mutation status was significantly associated with both OS ( p  = 0.005) and DFS (HR = 1.8; 95% CI = 1.3–2.3; p  = 0.0001).

However, it has also been suggested in the literature that KRAS is not suitable as an independent prognostic predictor for patients with CRC-PM treated with CRS. In a study conducted by Solomon et al ³⁸ that collected 100 patients who received CRS-HIPEC for CRC-PM, patients with mutant KRAS and WT patients (16 mo, 95% CI = 18.7–59.9 vs 13 mo, 95% CI = 11.1–20.8, p  = 0.594). In WT KRAS patients, no significant difference in median DFS after CRS-HIPEC was found when comparing patients who received and did not receive anti- EGFR therapy (10 mo, CI 95% 3.8, 16.6 vs 9 mo, 95% CI = 6.9–18.1, p  = 0.822). Graf et al ³⁹ performed KRAS mutation analysis in 110 of 399 subjects recruited who underwent CRS, and prognostic analysis similarly showed no difference in OS between KRAS -mutated tumors and KRAS -WT.

The above study showed that RAS is predictive of DFS in CRC patients after CRS treatment and can predict CRC-PM recurrence, but using RAS mutation status alone to predict OS in patients is controversial. Similar conflicting results have been seen using KRAS alone for prediction. Therefore, some studies have used the PSDSS ³⁴ or peritoneal carcinomatosis index (PCI) ³³ in combination with RAS to predict OS in patients and have shown good prognostic results.

BRAF

The BRAF oncogene is associated with more aggressive tumor biology and is a component of the RAS pathway. ^{40 41 42} The incidence of BRAF mutations in CRC-PM ranges between 10 and 20%, ^{9 43 44} and BRAF mutations are associated with an increased risk of peritoneal dissemination. ⁹

We summarize a series of studies in which various molecular indicators predicted the influence of CRS on the prognosis of CRC-PM patients. In the Kaplan–Meier analysis by Solomon et al, BRAF was the only mutated gene associated with poor DFS (16 mo, 95% CI = 11.7–43.3 vs 7 mo, 95% CI = 2.1–11.9, p  = 0.008). ³⁸ In multivariate analysis, m BRAF independently predicted early relapse ( p  = 0.032). ³⁸ According to Baratti et al, ³⁵ BRAF mutation (HR = 2.21, 95% CI = 1.05–4.63; p  = 0.038) and PCI (HR = 1.47, 95% CI = 1.03–2.10; p  = 0.036) were independently associated with poorer survival. Graf et al ³⁹ concluded that BRAF mutations were a marker of poor prognosis in patients with appendiceal and colorectal PM scheduled for CRS-HIPEC, and BRAF mutations ( p  = 0.0021) were significantly associated with poor prognosis. Tonello et al ³⁷ similarly came to the conclusion that m BRAF was significantly associated with both OS ( p  = 0.0171) and DFS ( p  = 0.001).

Several studies have shown a significant correlation between BRAF mutation status and the site of origin of CRC-PM operated with CRS. According to Baratti et al, ³⁵ right-sided primaries were associated with mutated KRAS ( p  = 0.01) and normal carcinoembryonic antigen (CEA; p  = 0.03). In another retrospective study of CRC-PM patients who underwent complete CRS, conducted at 16 different institutions and identifying a total of 796 patients, Péron et al ⁴⁵ found that right-sided CRC was more frequently associated with BRAF mutations and had microsatellite instability (MSI).

Because m BRAF tumors are more likely to be indistinguishable owing to MSI, BRAF mutations decrease DFS, OS, and cancer-specific survival in individuals with mCRC. Multiple studies have investigated the effect of BRAF and MSI on the therapy of CRC-PM patients. Larsen et al ⁴⁶ selected 257 patients with nonappendiceal PM-CRC treated with CRS-HIPEC, in which MSI was found in 29.3% of m BRAF cases. Patients with m BRAF /MSI had a higher 5-year survival rate compared with those with m BRAF combined with microsatellite stable (MSS) (58.3 vs 25.2%, p  = 0.022) and better 3-year DFS compared with m KRAS (48.6 vs 17.2%, p  = 0.049). The authors recommend pre-analysis of BRAF , RAS , and MSI status in patients with mCRC to tailor systemic therapy, and the potential prognostic value of these markers could be used to hel p select patients most suitable for CRS-HIPEC. ⁴⁶ Along with known clinical factors, the study by Tonello et al ³⁷ revealed that BRAF mutation ( p  = 0.01) and MS status ( p  = 0.04) were associated with survival in CRC-PM patients with CRS who had CRS.

The BRAF V600E mutation is the most common (80%) and aggressive m BRAF in CRC, identifying a subset of mCRC with a poorer prognosis. ⁴⁷ Results from the Flood et al ⁴⁷ analysis showed that BRAF V600E was associated with poorer OS (median: 28 mo, multivariate: HR = 2.29, p  = 0.026), DFS (median: 8 mo, multivariate: HR = 1.8, p  = 0.047), and postrecurrence survival (HR = 3.11, p  = 0.007) independently, and all CRC-PM patients with BRAF V600E mutations who had CRS recurred within 24 months. Because of the poor prognosis of patients with mCRC and the BRAF V600E mutation, Kopetz et al ⁴⁸ found that a combination of encorafenib, cetuximab, and binimetinib resulted in a significantly longer OS and a higher response rate than standard therapy in patients with mCRC with the BRAF V600E mutation.

However, another conflicting result emerged regarding the effect of m BRAF on survival in CRC-PM patients undergoing CRS. Larsen et al ⁴⁶ found a high prevalence of BRAF mutations (24%) but did not see any difference in survival after CRS-HIPEC. In another study that included 195 patients, Bhullar et al ³⁶ showed that BRAF ( p  = 0.109) mutation status did not predict OS in patients.

VEGF

Shaheen et al studied the effects of antibodies targeting the vascular endothelial growth factor ( VEGF ) receptor in a mouse model of CRC-PM and showed a 50% increase in survival in mice treated with anti- VEGF . ^{49 50 51} Shaheen et al ⁵² looked at how antibodies that target the VEGF receptor affected a mouse model of CRC-PM and found that mice given anti- VEGF lived 50% longer.

A study showed that intraperitoneal (IP) VEGF burden significantly increased ( p  = 0.013) and then gradually decreased ( p  < 0.005) after CRS, that neoadjuvant intravenous bevacizumab significantly decreased preoperative IP VEGF burden according to multifactorial analysis (OR = 5), that PCI significantly increased preoperative IP VEGF burden (OR = 10), that low preoperative IP VEGF load was associated with a significant increase in postoperative complications (OR = 4), and the neoadjuvant intravenous bevacizumab was the only factor affecting preoperative intravenous VEGF rates (OR = 1.7). ⁵³ In the study by de Cuba et al, ⁵⁴ high expression of VEGF was negatively associated with OS after CRS-HIPEC treatment ( p  = 0.02). In another analysis involving CRS-HIPEC in patients with PC, lower IP VEGF after incision (T1) was associated with improved OS ( p  = 0.0004) and DFS ( p  = 0.0006) at 2 years. ⁵⁵ A lower T1/IV VEGF ratio was also associated with improved OS ( p  = 0.004) and DFS ( p  = 0.0051). ⁵⁵

The results were similar in patients with CRC-PM. Logan-Collins et al ⁵⁶ studied tumor samples from patients receiving CRS-HIPEC for CRC-PM, and they reported that mean VEGF counts correlated with survival ( p  = 0.017), and this correlation was stronger for patients with recurrence ( p  = 0.002). Sluiter et al ⁵⁷ demonstrated that VEGF expression levels ( p  = 0.012) associated with OS in CRC-PM patients following CRS-HIPEC, ⁵⁷ with a mean survival of 38.3 months for those with low VEGF expression and 29.2 months for those with high VEGF expression ( p  = 0.035). Because versican (VCAN) -mediated angiogenesis may depend on the interaction with VEGF . ^{57 58 59} The authors, therefore, further focused on VCAN to assess the possible role of VCAN as a prognostic biomarker in CRC-PM patients treated with CRS-HIPEC and found that high epithelial VCAN expression was associated with more complete resection ( p  < 0.001), whereas high stromal VCAN expression was associated with better resection outcome ( p  = 0.003), VEGF expression levels ( p  = 0.012), and epithelial VCAN expression levels ( p  = 0.042) were associated with OS. ⁵⁷

Microsatellite Instability/Microsatellite Stable

Approximately 15% of CRC exhibit a hypermutated genotype accompanied by MSI-high (MSI-H) and defects in DNA mismatch repair. ^{60 61} We have delineated molecular subtypes of CRC by finding genomic events indicative of ultramutated, MSI-H/hypermutated, and MSS CRC. ^{61 62} MSI cases are crucial for diagnosis because two-thirds of MSI cases benefit from immune checkpoint inhibitors in mCRC. ⁶³

There is a substantial correlation between MSI or MSS status and the progression and prognosis of CRC-PM. Various investigations of CRS for CRC-PM, the following findings have been made. MSI patients exhibited a higher 5-year OS than MSS patients (58.3 vs 36.6%), whereas MSI/ WT patients had the greatest prognosis. ³⁷ MSS/WT and MSI/mutation patients exhibited comparable 5-year OS, whereas MSS/mutation patients had the worse prognosis ( p  = 0.0001). ³⁷ OS was related with MSI/WT (HR = 0.5) and MSS/WT-MSI/mutation (HR = 0.4), with a 5-year DFS of 62.5% for MSI/WT patients and 3.6% for MSS/mutation patients ( p  = 0.00001). ³⁷ Interestingly, one study identified a relationshi p between MSI and poor differentiation and the presence of embolism in the vascular system. ⁶⁴

MSI and MSS are frequently related to RAS and BRAF mutations for determining the prognosis of CRC-PM patients following CRS. A total of 13.3% of patients with MSI also possessed KRAS mutations, and 20.1% possessed BRAF mutations (no NRAS mutations). ³⁷ One report revealed that 29.3% of m BRAF cases included MSI. ⁴⁶ Patients with m BRAF /MSI had a greater 5-year survival rate than those with m BRAF combined with MSS (58.3 vs 25.2%, p  = 0.022) and a better 3-year DFS than those with m KRAS (48.6 vs 17.2%, p  = 0.049). ⁴⁶

There are several studies describing a less prominent role of MSI and MSS after CRC-PM who underwent CRS. For example, no significant difference in DFS was found when comparing patients diagnosed with MSI to those with MSS (17 mo, 95% CI = 9.4–28.3 vs 14 mo, 95% CI = 13.7–26, p  = 0.813). ³⁸ MSI cases did not have m BRAF status in Solomon et al's ³⁸ analysis, implying that the frequency of m BRAF in CRC-PM may be dependent on MSI status.

Considering the long-term clinical response and efficacy of checkpoint inhibition in MSI or MSS cancers in first-line therapy, ⁴⁷ MSI status is an essential element to consider when discussing the risks and advantages of CRS in multidisciplinary meetings.

APC

Loss of function or a buildu p of acquired mutations in the major driver gene adenomatous polyposis coli (APC) is responsible for the development of CRC. ⁶⁵ In the study by Baratti et al, ³⁵ phosphatidylinositol 3-kinase catalytic subunits p110α (PIK3CA) mutations were found in 17/78 patients (25.0%), and APC mutations were found in 25/68 patients (36.7%); APC mutations were univariately related with better survival ( p  = 0.03), and APC mutations were more prevalent in left-sided CRC. Yaeger et al ⁶² discovered a high enrichment of oncogenic alterations in APC and TP53 in left-sided initial tumors, and oncogenic abnormalities in APC were predictive of the survival status of CRC-PM following CRS (HR = 0.57, p  < 0.01).

PIK3CA

Mutations in the PI3K subunit PIK3CA contribute to oncogenesis and survival in a variety of solid tumors, and ∼10 to 20% of CRC carry PIK3CA activating mutations, making it one of the most commonly mutated genes in CRC and associated with advanced TNM stage, lower histological grade, OS, and progression-free survival (PFS). ^{66 67}

In a study including 53 consecutive CRC-PM patients who received optimal CRS-HIPEC, PI3K pathway alterations were an independent predictor of worse recurrence-free survival (RFS) (HR = 3.2, 95% CI = 1.3–8.3, p  = 0.01) with a clinically meaningful impact on median months to recurrence (5 vs 13 mo, p  = 0.02). ⁶⁸ Utilizing PIK3CA -targeted therapies to lower the probability of recurrence after optimum CRS, this innovative result may be used to customize therapeutic trials.

Serum Tumor Markers

Several studies have identified serum tumor markers (STM) such as CEA, cancer antigen 125 (CA125), carbohydrate antigen 19–9 (CA19–9), and α-fetoprotein (AFP) as indications of survival status and patient selection in CRC-PM patients treated with CRS. ^{69 70 71}

Huo et al ⁶⁹ showed that CEA > 6.5 mg/L and CA125 > 16 U/mL remained independent predictors of survival after adjusting for PCI [adjusted HR (aHR) = 2.46, 95% CI = 1.3–4.5, p  < 0.01 and aHR = 2.23, 95% CI = 1.21–4.09, p  < 0.01, respectively]. Patients with high CEA and low CA125 or vice versa had an increased risk of mortality (HR = 3.34, 95% CI = 1.21, 9.25, p  = 0.02; and HR = 2.76, 95% CI = 1.01, 7.42, p  = 0.04, respectively). ⁶⁹ High CEA levels combined with high CA125 levels had an additive impact, resulting in a 6-fold increase in mortality (HR = 6.57, 95% CI = 2.56; 13.69; p = 0.001, median survival: not attained vs 22 mo). ⁶⁹

Prognostic factors such as PCI play a key role in the decision-making process for surgical intervention. ^{47 72 73} However, the level of PCI that impacts the prognosis of patients with CRC-PM after CRS is still somewhat debatable. Péron et al, ⁴⁵ for instance, proposed that PCI ≤ 14 was an independent factor linked with improved OS and PFS. While Flood et al ⁴⁷ demonstrated that patients were offered CRS if their PCI was less than 15. Two studies by Hallam et al ^{74 75} found that a PCI of less than 12 was a negative prognostic factor. Therefore, several studies have attempted to integrate PCI with STM to provide a more accurate prognostic indicator. A low CEA/PCI ratio (<2.3) was associated with a longer median OS (56 vs 24 mo, p  = 0.001) and RFS (13 vs 9 mo, p  = 0.02) in 260 CRC-PM patients by CRS studied by Kozman et al. ⁷⁶ The influence of the CEA/PCI ratio on prognosis was most obvious in patients with PCI 10 (OS 72 mo vs 30 mo, p  < 0.001; RFS 21 vs 10 mo, p  = 0.002). In multivariate analysis, an elevated CEA/PCI ratio was independently related with a worse association between OS (corrected HR = 1.85, 95% CI = 1.13–3.10, p  = 0.02) and RFS (corrected HR = 1.58, 95% CI = 1.04–2.43, p  = 0.03). ⁷⁶

Clinically, STMs like as CEA, CA19–9, and CA125 are employed as tumor detection and diagnosis indicators. They are crucial for early cancer detection and prevention. Combining STM with other molecular markers in CRC-PM patients who needed CRS may be a viable strategy for the development of reliable predictors.

Others

In addition to the above molecules that are more commonly investigated in the treatment of CRC-PM patients by CRS, there are certain molecules that are less well documented; future study of these molecules may enhance the selection of CRC-PM CRS patients; therefore, additional description is required.

To investigate the changes in human heat shock protein 27 (Hsp27) in CRC-PM patients undergoing CRS-HIPEC, Kepenekian et al ⁷⁷ investigated Hsp27 mean expression levels at five time points: before CRS, end CRS, end HIPEC, 4 hours after HIPEC, and 24 hours after HIPEC. The high Hsp27 levels observed at the end of the procedure compared with before CRS ( p  < 0.0001) decreased during the HIPEC procedure but remained significantly higher than the levels measured before the procedure ( p  < 0.0005). Hsp27 levels decreased further after HIPEC and returned to baseline levels after 24 hours; in addition, there was a slight positive correlation between the increase in Hsp27 levels and the duration of CRS (0.43, p  = 0.05). ⁷⁷

Bloom syndrome protein (BLM) is a RecQ helicas that is primarily responsible for DNA unwinding. ⁷⁸ Recent studies have demonstrated that BLM protein expression is strongly associated with increased survival in CRS-HIPEC patients [high: 15 mo (11.3–18.7) vs low: NR; p  = 0.04]. ⁷⁹

Conclusion and Future Perspectives

A summary of the impact of various molecular states on the CRS for CRC-PM is shown in Table 1 . The current selection of patients with CRC-PM for CRS treatment relies on STM combined with medical imaging. However, due to the lack of sensitivity of CRC-PM imaging, ⁸⁰ PM may not be detected until surgical exploration, and the sensitivity and specificity of STM (CEA, CA125, etc.) used in clinical practice today to select patients for CRS are low. DNA (detection of mutations and methylation marks), RNA (mostly microRNA), proteins, and others have been identified as CRC-PM-related molecular biomarkers in recent years. Therefore, it is essential to learn more about how molecular state influences CRS in CRC-PM. Understanding all of the molecular events that occur in CRC-PM patients treated with CRS (e.g., RAS and BRAF mutations are associated with poorer DFS, shorter OS, and peritoneal recurrence) can aid in selecting individuals, guiding combination therapy, and predicting patient outcomes.

Table 1. Impact of molecular markers on cytoreductive surgery for colorectal cancer peritoneal metastases.

Molecular markers	Impact on CRS
RAS	RAS mutations, especially KRAS mutations, are linked to a shorter DFS, a worse OS, and peritoneal recurrence after CRS.
BRAF	BRAF mutations, especially BRAF V600E, were linked to a low DFS, a shorter OS, a primary site on the right side of the CRC-PM, and early recurrence after CRS.
VEGF	Low preoperative IP VEGF burden is associated with increased postoperative complications, and low postoperative VEGF expression is associated with improved OS and DFS.
VCAN	High epithelial VCAN expression is associated with more complete resection and better OS.
MSI/MSS	MSI patients had a longer survival than MSS patients, with the best prognosis for MSI/wild-type patients and the worst prognosis for MSS/mutant patients.
APC	APC mutations were associated with better survival and were more commonly mutated in left-sided CRC.
PIK3CA	PIK3CA is associated with an advanced TNM stage, a lower histological grade, OS, and PFS.
STM	A low CEA/PCI ratio was associated with a longer median OS and RFS.
Hsp27	At the end of the procedure, Hsp27 levels were higher than they were before CRS, and there was a link between higher Hsp27 levels and the length of CRS.
BLM	BLM protein expression was associated with an improved survival in patients after CRS.

Abbreviations: APC, adenomatous polyposis coli; BLM, Bloom syndrome protein; BRAF, b-viral oncogene homolog B1; CEA, carcinoembryonic antigen; CRC-PM, colorectal cancer peritoneal metastases; CRS, cytoreductive surgery; DFS, disease-free survival; Hsp27, heat shock protein 27; KRAS, Kirsten rat sarcoma viral oncogene homolog; MSI, microsatellite instability; MSS, microsatellite stable; OS, overall survival; PCI, peritoneal carcinomatosis index; PFS, progression-free survival; PIK3CA, phosphatidylinositol 3-kinase catalytic subunits p110α RAS, Rat sarcoma viral oncogene homolog; RFS, recurrence-free survival; STM, serum tumor marker; VCAN, versican; VEGF, vascular endothelial growth factor.

There are also some limitations to this paper. First, the findings of certain molecules continue to be contested, and additional large cohort studies with increased regional or population sample sizes are required. Second, there are still too few studies on some of the molecules; therefore, it is necessary to investigate in depth the mechanisms by which these molecules contribute to CRS in CRC-PM patients and to validate the effect of molecular status on CRS in various cohorts. Since CRS combined with HIPEC is a typical treatment for CRC-PM patients, the majority of patients in the study received both CRS-HIPEC, which may have influenced the findings on the impact of molecular markers in CRS. Since CRS combined with HIPEC is a typical treatment for CRC-PM patients, the majority of patients in the study received both CRS-HIPEC, which may have influenced the findings on the impact of molecular markers in CRS. To improve the efficacy of CRS treatment and promote the development of CRC-PM therapy, the study methods need to be further optimized and improved, the status of different molecules need to be further detected at different periods (e.g., before CRS, after CRS, and after HIPEC), and the molecular mechanisms acting at different periods need to be investigated in depth. With the maturation of molecular biology technology, the combined detection of several indicators can increase the efficacy of CRC-PM in CRS, and the accurate therapy at the molecular level will undoubtedly be a blessing for CRC-PM patients.

Funding Statement

Funding This study is supported by National Key Clinical Discipline, the Guangzhou Science and Technology Bureau, Guangzhou, China (2023A04J2245), and the Sixth Affiliated Hospital, Sun Yat-Sen University Clinical Research –‘1010’Program:1010CG(2022)-08.