Survival for patients with metastatic colon cancer underwent cytoreductive colectomy in the era of rapid development of anticancer drugs: A real-world analysis based on updated population dataset of 2004–2018

Objective: Metastatic colon cancer (mCC) poses a great threat to the survival of patients suffering from it. In the past decade, many clinical trials have been carried out to improve the prognosis of patients with mCC. Numerous treatments have emerged, and satisfactory efficacy has been demonstrated in randomized phase III trials in highly selective patients with mCC. Our present study aims to investigate whether these therapeutic advances can be reflected to the broader mCC patients who performed cytoreductive colectomy. Method: General and prognostic data for patients diagnosed with mCC who underwent cytoreductive colectomy between 2004–2018 were extracted from the Surveillance, Epidemiology, and End Results (SEER) database. Survival was analyzed using the Kaplan-Meier method and Cox proportional hazards model. The hazard ratio (HR) and its 95% confidence interval (CI) were used to evaluate the influence of risk factors on prognosis. Results: A total of 26,301 patients diagnosed with mCC treated with cytoreductive colectomy were included in this study. The median overall survival was 19 months (range, 17–23). The good prognosis was associated with patients diagnosed at the most recent year, younger age, non-black race, female, married, without previous history of malignancy, no second malignancy onset, descending/sigmoid/splenic flexure colon tumor, normal CEA levels at diagnosis, low primary tumor burden, T1/T2 stage, N0 stage, single organ metastasis, underwent surgical resection of synchronous distant metastatic lymph nodes or organs, a high number of lymph-node examinations, low positive lymph-node ratio and received adjuvant chemotherapy. The proportion of patients surviving for ≥24 months increased from 37% in 2004 to 44.2% in 2016 (p < 0.001), especially in ≤49 years patients [46.8% in 2004 to 57.8% in 2016 (p < 0.001)]. The percentage of patients who died within 3 months decreased between 2004 and 2018 (from 19.6% to 15.7%; p < 0.001). Conclusion: Over a span of 15 years, the long-term survival has improved in real-world mCC patients who were treated with cytoreductive colectomy, especially among younger patients. However, the median overall survival remains not substantial.


Introduction
The incidence of colorectal cancer is rising in recent years, accounting for approximately 10% of all cancers. It is currently the third most common malignancy in the world, and the second most common cause of cancer-related deaths (Siegel et al., 2020;Biller and Schrag, 2021;Siegel et al., 2022). About 20% of patients with colorectal cancer are found to have distant metastases at the time of diagnosis, and up to 50% of patients progress to metastatic disease during subsequent follow-up (van der Geest et al., 2015). Metastatic colorectal cancer has a poor prognosis (Biller and Schrag, 2021;Modest et al., 2022). Since metastatic colorectal cancer encompasses a broad spectrum of clinical diseases, its clinical prognosis varies widely among patients. Unresectable metastatic disease has an extremely poor prognosis without systemic therapy and a mean OS of 6-8 months (Kawai et al., 2021). The chemotherapy regimen of 5-fluorouracil (5-FU) combined with oxaliplatin or irinotecan has been the cornerstone of the treatments of metastatic colorectal cancer since the 1960s (Douillard et al., 2000;Saltz et al., 2000;Goldberg et al., 2004;Lee et al., 2019;Modest et al., 2022). At the same time, by adding drugs that target vascular endothelial growth factor (VEGF) (such as bevacizumab, ramucirumab, and ziv-aflibercept), and drugs that inhibit the epidermal growth factor receptor (EGFR) signaling pathway (such as cetuximab and panitumumab) significantly improved median survival in patients with metastatic colorectal cancer (Mody and Bekaii-Saab, 2018). In addition, immune checkpoint inhibitors (such as pembrolizumab, and nivolumab) have made rapid progress in the field of colorectal cancer and brought new changes to the treatment of advanced colorectal cancer (Overman et al., 2017;Ganesh et al., 2019;André et al., 2020;Andre et al., 2021;Modest et al., 2022).
Through multidisciplinary comprehensive treatment strategies, the progression of colorectal cancer can be better controlled and alleviated, and the survival rate of most tumors can be improved. However, the progress made in the treatment of metastatic colorectal cancer is mostly reflected in the results of randomized controlled clinical studies, and these studies have a process of highly selective screening of cases for case inclusion. It is worth investigating whether the progress in the treatment of metastatic colorectal cancer is reflected in the real world. The purpose of this study was to investigate the shortterm and long-term survival of patients with metastatic colon cancer (mCC), who underwent cytoreductive colectomy from 2004 to 2018. This will aid in determining whether the advances of metastatic colorectal cancer treatment from randomized controlled clinical studies are extended to the broader mCC population.

Statistical analyses
All continuous variables in this study were described as mean ± standard deviation (SD) if they conformed to a normal distribution, and were compared by Student's t-test. If the variables did not conform to the normal distribution, they were described as the median and interquartile range (IQR) and compared with Wilcoxon rank-sum test. Frequency (%) was used to represent the classification variables and the chisquare test for comparison. Kaplan-Meier method was used to calculate the survival rate of patients. Univariate and multivariate Cox proportional hazard model and Fine and Gray model were used to estimate the hazard ratios (HRs) and sub-distribution hazard ratios (sHRs) of total overall survival (OS) and cancer-specific survival (CSS), and their corresponding 95% confidence intervals, respectively. This was done to assess the impact of different covariates on OS and CSS, and to determine independent risk factors. Ordered logistic regression analysis was used to evaluate the impact of different covariates on early death risk, and the odds ratio (OR) of risk factors and its 95% CI were calculated. In addition, in order to evaluate the HR of age, tumor size, tumor deposits, number of lymph node examinations, number of positive lymph nodes, and rate of lymph nodepositive on OS risk, the restricted cubic spline curve was used to display these correlations. The non-linear test was carried out through the likelihood ratio test, and the log-likelihood of the model with linear term and the model with cubic spline term was compared. All p-values were bilateral, and a p-value of <0.05 was considered to be statistically significant. R statistical package (v. 4.2.0) was used for evaluations.  Figure 2 shows that age, tumor size, tumor deposits, number of regional lymph node tests, number of regional lymph node positives, and positive rate of regional lymph nodes at the time of diagnosis did not linearly correlate with the risk of all-cause deaths in patients with mCC (p for non-linear <0.05). These factors were then transformed into categorical variables and incorporated into the Cox model. After adjusting other covariates, they were all found to be independent risk factors for all-cause deaths (Table 3).

Short-and long-term survivors
We define short-term survivors as patients who died within 3 months of initial diagnosis, and long-term survivors as patients who survived for at least 24 months. Figure 4 shows that the proportion of short-term survivors in the total population decreased from 19.6% in 2004 to 13.9% in 2015 (p < 0.001). The proportion of long-term survivors increased from 62.7% in 2004 to 55.8% in 2016 (p < 0.001). In addition, survivors who survived for at least 12 months increased from 58.4% in 2004 to 67.0% in 2015. It Frontiers in Pharmacology frontiersin.org was found that the trend of survival improvement was more prominent in the younger population (Figures 3, 4). The good long-term survival was associated with patients diagnosed at the most recent year, younger age, non-black race, female, married, without previous history of malignancy, no second malignancy onset, descending/sigmoid/splenic flexure colon tumor, normal CEA levels at diagnosis, low primary tumor burden, T1/ T2 stage, N0 stage, single organ metastasis, underwent surgical resection of synchronous distant metastatic lymph nodes or organs, a high number of lymph-node examinations, low positive lymphnode ratio and received adjuvant chemotherapy (Table 4).

Discussion
The study mainly found that the median OS of patients with mCC, who underwent cytoreductive surgery, remained largely stable and not substantial (17-21 months) over a duration of 15 years, however, the risk of mortality for mCC patients who underwent cytoreductive surgery has decreased in recent years compared to the past. Besides, we focusing on patients at both extremes of the survival spectrum and found that the percentage of patients who died within 3 months decreased and the percentage of patients who survival more than 24 months increased, which fully affirms the survival benefits of colon cancer patients underwent cytoreductive surgery in the era of rapid development of anticancer drugs.
Since the 1990s, 5-FU in combination with leucovorin has become the mainstay of treatment for metastatic colorectal cancer, and studies have reported a median survival of approximately 12 months (Poon et al., 1989). Subsequent studies have evaluated the effects of FOLFOX (a combination of 5-FU, leucovorin, and oxaliplatin) and FOLFIRI (a combination of 5-FU, leucovorin, and irinotecan) for Frontiers in Pharmacology frontiersin.org metastatic colorectal cancer. These multidrug regimens resulted in a median OS of 12-20 months (de Gramont et al., 2000;Douillard et al., 2000;Saltz et al., 2000), suggesting that the multidrug combination may prolong patient survival. Therefore, FOLFOX and FOLFIRI regimens have become the standard firstline chemotherapy regimens for the treatment of metastatic colorectal cancer, and there is no significant difference in tumor benefit between the two regimens (Lee et al., 2019). The emergence of molecularly targeted drugs has further improved the treatment of metastatic colorectal cancer. Bevacizumab, a monoclonal antibody targeting VEGF, inhibits tumor angiogenesis in combination with chemotherapy. Hurwitz et al. (2004) showed that compared with FOLFIRI alone, FOLFIRI combined with bevacizumab significantly prolonged the OS of patients (15.6 months vs. 20.3 months). The use of FOLFOXIRI combined with bevacizumab in patients with unresectable metastatic colorectal cancer was shown to have a significant overall objective response rate (69%), with a respectable conversion rate of 40% in selected patients (Tomasello et al., 2017). Other targeted angiogenesis drugs, such as ziv-aflibercept and ramucirumab, are recommended in combination with FOLFIRI for second-line treatment of metastatic colorectal cancer, and their median progression-free survival and objective response rate are superior to FOLFIRI therapy alone (Lee et al., 2019). Another important target for metastatic colorectal cancer therapy is EGFR, a tyrosine kinase closely related to HER2/neu that is overexpressed in many tumors. It is involved in signaling pathways, such as tumor proliferation, angiogenesis, and migration. Cetuximab and panitumumab are EGFR inhibitors and were found to have a significant survival advantage over supportive care in patients with chemotherapy-refractory tumors, and when combined with chemotherapy as first-line therapy, it significantly improved PFS and OS (Riedesser et al., 2022). Currently, anti-EGFR therapy is mainly used for treating metastatic colorectal cancer patients with wild-type RAS/BRAF, since RAS/BRAF mutant patients do not respond well to anti-EGFR therapy.
The RAS gene is often mutated in metastatic colorectal cancer, and the most mutated gene is the Kirsten Ras (KRAS) gene. More than 40% of metastatic colorectal cancers have KRAS mutations (Ros et al., 2021;Ciardiello et al., 2022). RAS mutations are often associated with poor prognosis and decreased response to antitumor therapy. Therefore, treatments targeting this RAS gene mutation are beneficial for the majority of patients. Currently, there is a lack of effective drugs for the treatment of RAS gene mutations. Sotorasib and adagrasib selectively inhibit KRAS G12C mutation. Preliminary research results confirm their efficacy in KRAS G12C mutation patients, and follow-up research is still in progress (Hong et al., 2020;Weiss et al., 2021). BRAF mutations are present in approximately 10-15% of patients with metastatic colorectal cancer. The V600E mutation is the most common mutation in the BRAF gene. BRAF mutations abnormally activate the MAPK signaling pathway, making tumors highly aggressive. Patients with BRAF V600E-mutated metastatic colorectal cancer are poorly responsive to chemotherapy and have an Frontiers in Pharmacology frontiersin.org extremely poor prognosis with a median survival of only 12 months (Cohen et al., 2021). Some selective BRAF inhibitors have been developed, such as vemurafenib and dabrafenib. Unfortunately, these BRAF inhibitors negatively activate the MAPK signaling pathway, making their singleagent efficacy less than ideal. Combining multi-target blockade may help to obtain a more effective antitumor response. The BEACON study confirms that encorafenib plus cetuximab has significant advantages in objective response rate and OS in BRAF V600E-mutated metastatic colorectal cancer (Tabernero et al., 2021) In addition, studies combining encorafenib, cetuximab, and chemotherapy are also underway (Kopetz et al., 2021). Aberrant alterations in the HER2 gene are relatively uncommon in metastatic colorectal cancer, with HER2 gene amplification present in approximately 3% of metastatic colorectal cancers, mostly in RAS/BRAF wild-type patients (Ross et al., 2018). Retrospective studies have shown that HER2 gene amplification is associated with resistance to anti-EFGR therapy . Currently, a series of studies on anti-HER2 therapy drugs in RAS/BRAF WT metastatic colorectal cancer patients is underway, and encouraging results have been obtained. This indicates that targeting HER2 therapy has a strong potential for treating metastatic colorectal cancer Sartore-Bianchi et al., 2020;Siena et al., 2021). Immunotherapy has yielded favorable outcomes in patients with a variety of solid tumors. However, in metastatic colorectal cancer, only a small proportion of patients may benefit.   Microsatellite-instability-high (MSI-H)/mismatch repairdeficient (dMMR) is a biomarker for predicting the efficacy of immunotherapy in metastatic colorectal cancer, and it is present in approximately 5% of tumors in metastatic colorectal cancer (Arrichiello et al., 2021). Results of a multicenter randomized phase III clinical trial showed that compared with chemotherapy, pembrolizumab significantly prolonged median progression-free survival (PFS) in dMMR/MSI-H colorectal cancer patients, reaching 16.5 months. The majority of pembrolizumab monoclonal antibody-treated patients achieved objective responses over time (84% of patients lasting ≥2 years) (André et al., 2020). For microsatellite stable (MSS)/mismatch repair proficient (pMMR) patients with a high proportion of metastatic colorectal cancer, the efficacy of immunotherapy is poor, and whether it is suitable to receive immunotherapy is still under investigation. Studies have shown that cytotoxic drugs, anti-angiogenic drugs, molecularly targeted therapy, and radiotherapy can activate immunogenic cell death in tumor cells. Therefore, a series of studies are investigating the potential role of immune checkpoint inhibitor combination therapy (Ciardiello et al., 2022).  Frontiers in Pharmacology frontiersin.org Reduction in the early death of cancer patients is an important indicator to evaluate the effect of comprehensive cancer treatment and nursing. McPhail et al. (2015) analyzed early mortality in patients with breast, colorectal, lung, prostate, and ovarian cancer in the United Kingdom, and found that age, tumor stage at diagnosis, income, and geographic location were significantly associated with early mortality in colorectal cancer. An analysis of colorectal cancer patients in the United Kingdom between 2006 and 2008 found that around 11.5% of colon cancer patients died within a month of being diagnosed, and about 33% died within a year of being diagnosed. Of these, old age, late tumor stage, poverty, and visits to the emergency department were associated with early mortality (Downing et al., 2013). Lieu et al. (2014) found that age was significantly associated with OS within 1 year of diagnosis in patients with metastatic colorectal cancer. The age effect was U-shaped, wherein both, young and old age, were unfavorable factors for early mortality in patients with metastatic colorectal cancer. Younger patients are generally healthier and have fewer underlying diseases than older patients, but their OS within 1 year of diagnosis is suboptimal, which may suggest different tumor biology. Renfro et al. (2016) found that low BMI was associated with an increased risk of metastatic colorectal cancer progression and death. Metastatic colorectal cancer patients are prone to cachexia, which leads to a significant decrease in BMI, affects the subsequent treatment of patients, and results in a significant increase in the risk of death.
A pooled analysis of 9 clinical trials examined the impact of performance status (PS) on chemotherapy in patients with metastatic colorectal cancer (Sargent et al., 2009). Although patients with PS grade 2 achieved similar treatment benefits as those with PS grades 0 and 1, they had significantly higher 60-day mortality (12.0% vs. 2.8%, p < 0.001). This suggests that performance status is an important risk factor for early mortality in patients with metastatic colorectal cancer. In a phase III randomized controlled trial of irinotecan in metastatic colorectal cancer, Giessen et al. (2011) assessed the clinicopathological factors associated with 60-day mortality in patients. The 60-day mortality rate in the study was 5.0% (24/ 479). Elevated LDH and WBC levels were considered independent predictors for early mortality, and fitness status showed a negative trend with an increased risk of early mortality in the study, but it was not statistically significant. However, in another study by Giessen et al. (2013). WBC count and performance status were identified as significant risk factors for early mortality, whereas LDH levels were not found to be associated with early mortality. In a randomized controlled study of primary tumor resection combined with systemic therapy (van der Kruijssen et al., 2021), patients with metastatic colorectal      cancer who received systemic therapy after primary tumor resection had significantly higher 60-day mortality than patients who received systemic therapy alone. In the surgical group, factors such as serum lactate dehydrogenase, aspartate aminotransferase, alanine aminotransferase, and neutrophil count were associated with 60-day mortality.
In addition, the 60-day mortality appears to be significantly high in patients with multiple risk factors before surgery. There may be multiple reasons for how these factors affect early mortality in patients. 1) These risk factors are considered indicators of tumor mutational burden and malignancy (Kleespies et al., 2009;Ahn et al., 2014;Turner et al., 2015). Elevated levels of multiple indicators in patients suggest a large tumor burden, and therefore patients are at a risk of rapid progression after surgery. 2) These biochemical markers have been associated with poor prognosis in multiple studies (Ahn et al., 2014;Li et al., 2016). 3) Elevated levels of neutrophils in patients suggest a more pronounced systemic inflammatory response, which significantly increases the mortality of patients (Dell'Aquila et al., 2018). Early mortality was analyzed in a pooled analysis of 28 randomized clinical trials of metastatic colorectal cancer, which collected data from more than 22,000 patients with metastatic colorectal cancer from the ARCAD database. In this analysis, early mortality at 30 days, 60 days, and 90 days was 1.4%, 3.4%, and 5.5%, respectively. Older age, lower BMI, poor performance status, multiple metastatic sites, BRAF mutation status, and elevated laboratory markers (elevated bilirubin, WBC count, and neutrophil count) are associated with 90day death. In contrast, KRAS mutation status, sex, single metastases, primary tumor site, and prior chemotherapy and treatment class (targeted versus non-targeted) were not associated with early mortality (Renfro et al., 2017). For patients with metastatic colorectal cancer with multiple risk factors, treatment options should be carefully evaluated to improve patient survival.
Our study has several limitations. First, the study was retrospective and the analysis relied on administrative claims data, with the potential for misclassification of cancer stage, vital status, and cause-specific survival. Second, the research object of this article is patients with mCC who underwent cytoreductive colectomy, it had clumped together a very heterogenous disease into a single category, which may be misleading. For example, the cohort will include patients with oligometastatic disease who underwent curative surgery and patients with bowel obstruction who underwent surgery, one has good outcomes, while the other is of poor prognosis. While, considering the different severity of metastatic colon cancer corresponding to different cytoreductive colectomy type and had a different survival outcome, we had classified the types of cytoreductive colectomy (Subtotal colectomy/hemicolectomy, Total colectomy/ proctocolectomy, Partial colectomy/segmental/Local excision) and included into multivariate analysis. These type of cytoreductive colectomy are common in clinical practice for metastatic colon cancer patients. Third, there is no information on patient Karnofsky performance status, comorbidities, drug use of chemotherapy and targeted immunotherapy, or information on gene mutation, and imaging information (computerized tomography, magnetic resonance imaging, Positron Emission Tomography-Computed Tomography and so on) in the SEER database. However, the SEER database has done the following to ensure data accuracy: populationbased case identification, detailed review of medical and pathological records, strict data collection and quality control standards, and high Respectively adjusted for the covariables of age at diagnosis, race, sex, marital status, previous tumor history, lifetime number of tumors, site of the tumor, CEA, level, size of the tumor, histology, T stage, regional lymphnodes examined, surgery type, surgery for distant metastasis organ and lymphnode site, radiation therapy, and chemotherapy therapy. d Respectively adjusted for the covariables of age at diagnosis, race, sex, marital status, previous tumor history, lifetime number of tumors, site of the tumor, CEA, level, size of the tumor, histology, T and N stage, regional lymphnodes examined, surgery type, surgery for distant metastasis organ and lymphnode site, radiation therapy, and chemotherapy therapy based on the dataset of 2010-2018 which included the information of perineural invasion and number of tumor deposits.
Frontiers in Pharmacology frontiersin.org patient follow-up rates. This data has been thoroughly audited for accuracy and completeness. Therefore, it is reasonable to assume that our data on patient characteristics, tumor pathology and staging, treatment modalities, and survival status are reasonably accurate, and even with limited data, our conclusions remain reasonable. In addition, it should be noted that this study mainly included metastatic colon cancer patients who had undergone cytoreductive surgery and did not study rectal cancer, so the conclusions are not applicable to the metastatic rectal cancer population.

Conclusion
Over a span of 15 years, the OS and long-term survival of patients with mCC improved slightly, especially in the younger patient population. This demonstrates the progress in the comprehensive treatment for mCC over the past few decades. At the same time, we need to recognize that there is still a lot of room for improvement in the future.

Data availability statement
The original contributions presented in the study are included in the article/Supplementary Material, further inquiries can be directed to the corresponding author.

Author contributions
FC and GM conceived the research and wrote the manuscript. FC and SY analysed the data and prepared the figures and tables. All authors were involved in revising the manuscript and have approved its final version.