Abstract
Background
The impact of heat shock proteins (HSPs) expression levels on the prognosis of patients with colorectal cancer (CRC) remains controversial. Therefore, we conducted a meta-analysis to evaluate the prognostic significance of HSPs expression in CRC patients and its relationship with clinical pathological features.
Methods
We searched for eligible studies from PubMed, Embase, Cochrane, and Web of Science databases, and extracted the hazard ratios (HRs) and 95% confidence intervals (CIs) for overall survival (OS), disease-free survival (DFS), progression-free survival (PFS), relapse-free survival (RFS), or cancer-specific survival (CSS), along with their clinical pathological characteristics. Meta-analysis was performed using either a fixed-effects model or a random-effects model, depending on the heterogeneity between studies.
Results
A total of 20 studies, including 3919 CRC patients, were included in this meta-analysis. The results showed that high expression of HSPs can predict poor OS in CRC patients (HR = 1.79, 95% CI: 1.39–2.30, P < 0.00001). Subgroup analyses revealed that high expression of HSP70 (HR = 1.75, 95% CI: 1.10–2.78, P = 0.02), HSP27 (HR = 1.73, 95% CI: 1.34–2.25, P < 0.0001), and HSP47 (HR = 2.91, 95% CI: 1.39–6.08, P = 0.005) were associated with increased risk for poorer OS, but the expression of HSP90 (HR = 1.28, 95% CI: 0.60–2.72, P = 0.52) was not significantly correlated with OS in CRC patients. Moreover, high expression of HSPs may predict poor DFS (HR = 1.56, 95% CI: 1.18–2.06, P = 0.002), RFS (HR = 2.15, 95% CI: 1.37–3.38, P = 0.0008), and CSS (HR = 1.65, 95% CI: 1.07–2.42, P = 0.0003) in CRC patients. In addition, high expression of HSPs was significantly correlated with high TNM stage (OR = 1.29, 95% CI: 1.01–1.65, P = 0.04).
Conclusion
These findings indicate that high expression of HSP70, HSP27, and HSP47 may have a significant adverse effect on the prognosis of CRC patients, and contribute to the prognosis evaluation of these patients and potentially guide the development of targeted therapeutic strategies.
Supplementary Information
The online version contains supplementary material available at 10.1186/s12957-026-04239-0.
Keywords: Heat shock proteins, Colorectal cancer, Prognostic value, Clinicopathological significance, Meta-analysis
Introduction
Colorectal cancer (CRC) is a common malignant tumor in the gastrointestinal tract. According to the World Health Organization’s International Agency for Research on Cancer (IARC) team, the number of new cases of CRC worldwide in 2022 was approximately 1.926 million, ranking third; the death toll is nearly 904,000, ranking second only to lung cancer [1]. Although surgery, radiotherapy, chemotherapy, targeted therapy, and immunotherapy can achieve certain clinical effects and improve the 5-year survival rate of CRC patients, tumor heterogeneity still affects the treatment effectiveness and prognosis of CRC, and the recurrence rate and long-term survival rate remain suboptimal [2, 3]. Therefore, in recent years, increasing studies have begun to focus on the important role of CRC prognostic predictive markers in the clinic. Finding prognostic markers to accurately predict the prognosis of CRC can assist clinicians in formulating a reasonable therapeutic plan and achieving optimal therapeutic outcomes, which is of great clinical significance.
Heat shock proteins (HSPs) are the most representative molecular chaperones that play a crucial role in stabilizing protein structures, promoting the repair or degradation of damaged proteins, and maintaining protein homeostasis and cellular function [4]. According to molecular weight, HSPs can be classified into HSP27, HSP40, HSP60, HSP70, HSP90, and HSP110 [5]. Studies have shown that HSPs are closely related to the molecular mechanisms of tumor occurrence and development, promoting the proliferation, invasion, and metastasis of cancer cells [6]. HSPs are highly expressed in a variety of malignant tumors, including lung cancer, colorectal cancer, gastric cancer, esophageal cancer, breast cancer, and liver cancer, and are recognized as oncoproteins in various tumor types [7–11]. Their overexpression is associated with tumor survival, metastasis, and anticancer drug resistance, directly affecting the prognosis of tumors [6].
In recent years, numerous reports have explored the clinical and prognostic significance of expression of HSPs in CRC patients, but their exact role remains controversial. Moreover, different types of HSPs may exhibit different prognostic outcomes, and there is currently a ack of comprehensive meta-analysis integrating multiple HSP subtypes to assess their impact on various survival outcomes and key clinicopathological characteristics. Drecoll et al. [12] found that patients with high expression of HSP90 in CRC had better survival and that this was significantly correlated with tumor infiltration depth and lymph node metastasis, though no association was found between HSP90 levels and other clinical pathological features. Bauer et al. [13] reported opposite results, indicating that high expression of HSP27 was an independent risk factor for poor prognosis in CRC, with no correlation observed between high expression of HSP27 and tumor infiltration depth, lymph node metastasis, distant metastasis, or tumor staging. Therefore, we updated the relevant literature search, conducted a comprehensive and integrated meta-analysis of multiple HSP subtypes, and further explored the prognostic value of HSPs expression in CRC patients and its detailed association with major clinical pathological features.
Methods
Search strategy
Two investigators were responsible for crafting the search strategy. They independently developed subject terms and keywords for the search of multiple databases, including PubMed, Embase, the Cochrane Library, and Web of Science, covering the period from the establishment of the databases to 10 November 2025. The search utilized a wide range of terms, such as “colorectal cancer,” “colorectal neoplasm,” “colorectal tumor,” “Colorectal Carcinoma,” “Heat shock proteins,” “Heat-Shock Proteins,” “Stress Protein,” “HSP,” and others. The references of the included studies were also screened for further information. Supplementary File S1 presents the literature search strategy. The reporting of this study followed the guidelines outlined in the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA 2020) statement [14], and the research protocol was registered on the International Prospective Systematic Evaluation Registry (PROSPERO: CRD42024561226, available from https://www.crd.york.ac.uk/PROSPERO/view/CRD42024561226).
Selection criteria
The inclusion criteria are as follows: (1) the original studies that were prospective or retrospective cohort studies and clinical trials; (2) studies that assessed the prognostic impact of HSPs on overall survival (OS), disease-free survival (DFS), progression-free survival (PFS), event-free survival (EFS), relapse-free survival (RFS), or cancer-specific survival (CSS) in patients with colorectal cancer; (3) studies provided data on the hazard ratio (HR) with a 95% confidence interval (CI), which could be either extracted directly from the studies or computed based on available data; (4) papers published in English; and (5) studies that have been fully published.
The exclusion criteria are as follows: (1) review articles, case reports, meeting abstracts, comments, letters, re-analyses, and animal studies; (2) studies lacking sufficient information to compute HRs and 95% CIs; (3) studies not providing survival data; and (4) studies with overlapping or duplicated data.
The literature search and screening was done independently by two researchers (Ziqing Wang and Zhenmin Zhang) based on the search criteria and inclusion and exclusion criteria. The search results were compared, and any inconsistencies were resolved by discussion or arbitration by a third researcher (Jixuan Zou).
Data extraction
Two reviewers (Ziqing Wang and Zhenmin Zhang) independently used standardized forms to extract relevant data from the included studies. The following data were extracted: the first author’s name, year of publication, number of patients, study region, age and gender of patients, sample type, disease type and stage, cut-off value, type of HSPs, detection methods, follow-up duration, number of patients with lymph node and distant metastases, infiltration depth grades, and HRs and 95% CIs for OS, DFS, PFS, and RFS. In multivariate analysis, there may be multiple confounding factors associated with the study outcomes. By incorporating these confounding variables, the adjusted model can more accurately estimate the independent association between exposure factors and outcomes, thereby reducing confounding bias. Therefore, when an article presents multiple HRs derived from both univariate and multivariate analyses, the result stemming from the multivariate analysis takes precedence, considering its potential for greater accuracy.
Quality assessment
The Newcastle-Ottawa Quality Assessment Scale (NOS) [15] was used to evaluate the methodological quality of each included study. The NOS scale comprises three dimensions: selection, comparability, and outcome, and awards a maximum score of nine. Articles achieving a final score of six or above were deemed to be of high quality.
Statistical analysis
The statistical analyses were conducted utilizing STATA version 14.0 (Stata Corporation, College Station, TX, USA) and Review Manager version 5.3 (the Cochrane Collaboration, Oxford, UK). A P value of less than 0.05 on both sides was considered statistically significant. HRs and corresponding 95% CIs were used to assess the prognostic impact of HSPs in CRC patients for various survival outcomes (OS, DFS, etc.). Among them, DFS, PFS, and EFS, although used as proprietary terms for different treatment methods or clinical trials, but their definitions were essentially similar, which was the time from randomization to disease progression or death from any cause. Considering that there were relatively few reports on them retrieved in this study, a combined analysis was conducted. Furthermore, compared to CRC patients with low expression of HSPs, HR > 1 was indicated a poorer prognosis in CRC patients with high expression of HSPs. The results for dichotomous variables were reported using the odds ratio (OR) and its 95% CI. The heterogeneity among the included studies was assessed using the Q test and I² statistic. If the P-value ≥ 0.1, or if the I² ≤ 50%, the heterogeneity was considered to be not statistically significant, prompting the use of a fixed effect model for analysis. Conversely, if these criteria were not met, the heterogeneity was considered statistically significant. Additional sensitivity or subgroup analyses were conducted to identify the source of heterogeneity. If heterogeneity persisted, a random effect model would be employed for the meta-analysis. We conducted sensitivity analysis to examine the impact of each study on the overall hazard ratio (HR). To evaluate for publication bias, funnel plots were utilized along with Egger’s and Begg’s tests, with a P-value < 0.05 indicating the presence of such bias.
Results
Study selection
As shown in Fig. 1, a total of 1,176 articles related to the aforementioned search terms were retrieved. After removing duplicate articles and applying the inclusion and exclusion criteria, 124 studies were obtained, following the exclusion of reviews, conference abstracts, systematic evaluations, and studies irrelevant to the topic based on title and abstract review. After full-text reading, 104 articles were excluded due to a lack of available outcome indicators, insufficient data, or overlapping data. Finally, this meta-analysis encompassed 20 studies that fulfilled the criteria.
Fig. 1.
Flow diagram of study selection
Characteristics of included studies
This study included 20 articles [12, 13, 16–33], involving 3,919 CRC patients, with sample sizes ranging from 51 to 456. Twenty studies evaluated seven different types of HSPs, including HSP70 (7) [13, 16–18, 22, 23, 27], HSP27 (5) [13, 28–31], HSP90 (3) [12, 32, 33], HSP110 (3) [16, 21, 26], HSP47 (2) [24, 25], HSP60 (1) [20], and HSPB3 (1) [19]. One article evaluated the impact of HSP70 and HSP110 on the prognosis of colorectal cancer patients [16], while another assessed the impact of HSP70 and HSP27 [13]. The studies encompassed patients from various countries, including four from China, four from Hungary, three from Germany, three from South Korea, and two each from the UK, Greece, and Japan. All NOS scores of the included studies were ≥ 6, indicating high quality. Table 1 presents the characteristics of the included studies.
Table 1.
Summary of the data extracted from the 20 studies included
| First author (Year) |
Country | Patient number (M/F) |
Median/mean age (years) | Sample type | Cancer type |
Disease stage |
HSPs type | Rate Positive/high [n (%)] |
Cut-off value | Criteria for Cut-off value | Detection method |
Median/mean/ range follow-up (months) |
Outcome | Multivariate adjustment | Quality score |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Banerjea (2005)[16] | Britain | 52 (26/26) | NA | tumor tissue | MSI-H CRC | Ⅰ-Ⅳ | HSP70 | 47 90.4%) | 30% | IHC-based | IHC | 0.96–59.3 | OS | Yes | 7 |
| HSP110 | NA | 30% | |||||||||||||
| Bauer (2012)[13] | Germany | 355 (200/155) | 66 (25–91) | tumor tissue | colon cancer | Ⅰ-Ⅳ | HSP27 | 103 (48.4%) | IHC score = 3 a | IHC-based | IHC | 87 ± 27.6 | OS | Yes | 7 |
| HSP70 | 141 (39.9%) | IHC score = 4 a | |||||||||||||
| HSP60 | 219 (62.4%) | IHC score = 8 a | |||||||||||||
| Drecoll (2014)[12] | Germany | 355 (200/155) | 66 (25–91) | tumor tissue | colon cancer | Ⅰ-Ⅳ | HSP90 | 105 (29.6%) | IHC score = 4 a | IHC-based | IHC | 87 ± 27.6 | OS | Yes | 7 |
| Gráf (2018)[17] | Hungary | 232 (138/94) | 66.82 (11.41) | serum | CRC | Ⅰ-Ⅳ | HSP70 | NA | 1.64 ng/ml | Serum concentration-based | ELISA | 46.42 (1–60) | OS | Yes | 8 |
| Jubran (2017)[18] | Hungary | 235 (139/96) | 66.6 (mean) | serum | CRC | Ⅰ-Ⅳ | HSP70 | 110 (46.8%) | 1.65 ng/ml | Serum concentration-based | ELISA | 1–60 | OS | No | 7 |
| Kalioraki (2020)[19] | Greece | 188 (98/90) | 68 (35–93) | tumor tissue | CRC | Ⅰ-Ⅳ | HSPB3 | 137 (72.9%) | 4.20 RQU(27%) | Serum concentration-based | RT-PCR | 1-100 | OS; RFS | Yes | 8 |
| Kang (2023)[20] | Korea | 456 (272/184) | 65.0 ± 11.4 | tumor tissue | CRC | Ⅰ-Ⅳ | HSP60 | 214 (42.1%) | IHC score = 3 b | IHC-based | IHC | 1–60 | CSS; EFS | Yes | 8 |
| Kim (2014)[21] | Korea | 168 (79/89) | NA | tumor tissue | MSI-H CRC | Ⅰ-Ⅳ | HSP110 | 128 (76.2%) | IHC score = 2 b | IHC-based | IHC | 56.1 (1-108.7) | DFS | Yes | 8 |
| Kocsis (2010)[22] | Hungary | 175 (99/76) | 66.3 ± 10.7 | serum | CRC | Ⅰ-Ⅳ | HSP70 | NA | 1.67 ng/ml | Serum concentration-based | ELISA | 33.0 (23.5–43.5) | OS | No | 7 |
| Lazaris (1995)[23] | Greece | 128 (63/65) | 65.72 (18–81) | tumor tissue | CRC | NA | HSP70 | 99 (77.3%) | 10% | IHC-based | IHC | 1–60 | OS | No | 7 |
| Mori (2020)[24] | Japan | 139 (79/60) | 67 (mean) | tumor tissue | CRC | Ⅰ-Ⅳ | HSP47 | 69 (49.6%) | Gene level = 0.0969762 | Serum concentration-based | RT-PCR | 32.9 (0.4–115) | OS | Yes | 8 |
| Mori (2017)[25] | Japan | 157 (90/67) | 67 (35–89) | tumor tissue | CRC | Ⅰ-Ⅲ | HSP47 | 104 (66.2%) | IHC score = 4 c | IHC-based | IHC | 20.7 (1–62) | OS; DFS |
OS: No DFS: Yes |
8 |
| Oh (2017)[26] | Korea | 164 (83/81) | NA | tumor tissue | MSI-H CRC | Ⅰ-Ⅳ | HSP110 | 133 (81.1%) | H-score 208 d | IHC-based | IHC | NA | DFS | Yes | 8 |
| Rozenberg (2013)[27] | Hungary | 121 | NA | serum | CRC | Ⅰ-Ⅳ | HSP70 | NA | 1.65 ng/ml | Serum concentration-based | ELISA | 1–42 | OS | Yes | 8 |
| Schweiger (2015)[28] | Germany | 51 (29/22) | 63 (33–83) | tumor tissue | CRC | Ⅳ | HSP27 | 17 (33.3%) | 50% | IHC-based | IHC | NA | OS; RFS | No | 6 |
| Tweedle (2010)[29] | Britain | 404 (251/153) | NA | tumor tissue | CRC | Ⅰ-Ⅳ | HSP27 | 176 (43.6%) | IHC score = 3 e | IHC-based | IHC | 49.5 (0.23–146.5) | CSS | Yes | 8 |
| Wang (2012)[30] | China | 175 (98/77) | NA | tumor tissue | CRC | Ⅰ-Ⅳ | HSP27 | 85 (48.6%) | 70% | IHC-based | IHC | NA | OS | Yes | 7 |
| Yu (2010)[31] | China | 182 (105/77) | 61 (20–87) | tumor tissue | CRC | Ⅰ-Ⅳ | HSP27 | 86 (47.3%) | IHC score = 3 b | IHC-based | IHC | 1–60 | OS | Yes | 8 |
| Zhang (2020)[32] | China | 83 (52/31) | NA | serum | CRC | Ⅰ-Ⅳ | HSP90 | NA | NA | Serum concentration-based | ELISA | 10.3 (0.8–23.1) | PFS | Yes | 8 |
| Zhang (2019)[33] | China | 99 (54/45) | NA | tumor tissue | CRC | Ⅰ-Ⅳ | HSP90 | 55 (55.6%) | IHC score = 4 f | IHC-based | IHC | NA | OS | Yes | 7 |
M male, F female, NA not available, CRC colorectal cancer, OS overall survival, RFS relapse free survival, CSS cancer specific survival, EFS event free survival, DFS disease free furvival, PFS progression free survival
a Staining intensity was scored as 0 (none), 1 (weak), 2 (moderate), or 3 (strong). The staining percentage was scored as 0 (0), 1 (< 10%), 2 (10–50%), 3 (51–80%), and 4 (> 80%). Intensity × percentage=staining score (0–12)
b Staining intensity was scored as 0 (none), 1 (weak), 2 (moderate), or 3 (strong)
c Staining intensity was scored as 0 (none), 1 (weak), 2 (moderate), or 3 (strong). The staining percentage was scored as 0 (0), 1 (1–25%), 2 (26–50%), 3 (51–75%), and 4 (76–100%). Intensity × percentage=staining score (0–12)
d The H-score of each tumor was calculated as follows: [1× (percentage of area of score 1)] + [2 × (percentage of area of score 2)] + [3 × (percentage of area of score 3)]. Score 1: weaker nuclear-to-cytoplasmic expression than internal positive control; Score 2: nuclear-to-cytoplasmic expression intensity comparable to that of internal positive control; Score 3, stronger nuclear-to-cytoplasmic expression than internal positive control
e Staining intensity was scored as 0 (none), 1 (weak), 2 (moderate), or 3 (strong). The staining percentage was scored as 0 (0), 1 (< 5%), 2 (5–30%), 3 (31–70%), and 4 (> 70%). Intensity × percentage=staining score (0–12)
f Staining intensity was scored as 0 (none), 1 (weak), 2 (moderate), or 3 (strong). The staining percentage was scored as 0 (0), 1 (< 1/10), 2 (1/10–1/3), 3 (1/3–2/3), and 4 (≥ 2/3). Intensity × percentage=staining score (0–12)
Impact of HSPs on the OS in CRC patients
Fifteen articles have reported the relationship between the expression of HSPs and OS in CRC patients, involving a total of 2,644 patients. Among them, two articles [13, 16] examined two types of HSPs. Due to substantial heterogeneity among the studies (I2 = 68%, P < 0.0001), a random effect model was adopted (Fig. 2). The overall HR for the OS was 1.79 (95% CI: 1.39–2.30, P < 0.00001) in CRC patients with high HSPs expression compared with those with low HSPs expression (Fig. 2), indicating that high HSPs expression could predict poorer OS in CRC patients.
Fig. 2.
Forest plots for the association between the expression of HSPs and OS in CRC patients
The results of the meta-analysis of the effect of HSPs on OS in CRC patients showed that 15 studies have significant heterogeneity (I2 = 68%, P < 0.0001). To account for the possibility that different types of HSPs could influence the results and contribute to heterogeneity, we conducted a subgroup analysis on the impact of different types of HSPs on OS in CRC patients. As shown in Fig. 3; Table 2, high expression of HSP70, HSP27, and HSP47 was significantly associated with OS in CRC patients. The overall HR was 1.75 for HSP70 (95% CI: 1.10–2.78, P = 0.02), 1.73 for HSP27 (95% CI: 1.34–2.25, P < 0.0001), and 2.91 for HSP47 (95% CI: 1.39–6.08, P = 0.005). These results indicate that high expression of HSP70, HSP27, and HSP47 could predict poor OS in CRC patients. However, the overall results for HSP90 (HR = 1.28, 95% CI: 0.60–2.72, P = 0.52) showed no significant correlation between its expression and OS in CRC patients (Fig. 3).
Fig. 3.
Subgroup analysis of the relationship between different HSP expression and OS
Table 2.
Subgroup analysis of prognostic value of HSPs for OS in CRC patients
| Subgroups | No. of studies | No. of patients | Effects model | HR (95%CI) | p | Heterogeneity | |
|---|---|---|---|---|---|---|---|
| I2 (%) | p | ||||||
| OS | |||||||
| Total | 17 | 2644 | Random | 1.79 (1.39–2.30) | < 0.00001 | 68 | < 0.0001 |
| HSPs subtypes | |||||||
| HSP70 | 7 | 1298 | Random | 1.75 (1.10–2.78) | 0.02 | 80 | < 0.0001 |
| HSP27 | 4 | 763 | Random | 1.73 (1.34–2.25) | < 0.0001 | 2 | 0.38 |
| HSP90 | 2 | 454 | Random | 1.28 (0.60–2.72) | 0.52 | 80 | 0.03 |
| HSP47 | 2 | 296 | Random | 2.91 (1.39–6.08) | 0.005 | 0 | 0.77 |
| Region | |||||||
| Europe | 12 | 2299 | Random | 1.65 (1.20, 2.28) | 0.002 | 75 | < 0.00001 |
| Asia | 5 | 752 | Random | 2.18 (1.64, 2.88) | < 0.00001 | 0 | 0.92 |
| Sample size | |||||||
| n ≤ 200 | 12 | 1519 | Random | 1.99 (1.38, 2.85) | 0.0002 | 67 | 0.0004 |
| n > 200 | 5 | 1532 | Random | 1.51 (1.11, 2.05) | 0.008 | 64 | 0.02 |
| Sample type | |||||||
| tumor tissue | 13 | 2288 | Random | 1.65 (1.20, 2.26) | 0.002 | 71 | < 0.0001 |
| serum | 4 | 763 | Random | 2.16 (1.68, 2.77) | < 0.00001 | 0 | 0.44 |
| Detection method | |||||||
| IHC | 11 | 1943 | Random | 1.50 (1.07, 2.09) | 0.02 | 72 | 0.0001 |
| ELISA | 4 | 763 | Random | 2.16 (1.68, 2.77) | < 0.00001 | 0 | 0.44 |
| RT-PCR | 2 | 345 | Random | 3.13 (1.71, 5.72) | 0.0002 | 0 | 0.65 |
Different types of HSPs cannot explain the source of heterogeneity, therefore, we continue to conduct subgroup analysis in terms of region, sample size, sample type, and detection methods to further examine heterogeneity. Although we observed in each subgroup that CRC patients with high HSPs expression had poorer OS compared to those with low HSPs expression (Table 2), the results of subgroup analysis indicate that heterogeneity cannot be explained by region, sample size, sample type, and detection method (Table 2). Next, we examined heterogeneity by removing one study at a time and found that when the data from Banerjee et al. (2005) was removed, the heterogeneity of the remaining studies decreased significantly compared to before (I2 = 49%), indicating that this study was the source of heterogeneity. After further analysis of this study, we thought it was because this study only included MSI-H CRC patients. After removing this study, the pooled HR was 1.94 (95% CI: 1.58–2.37, I2 = 49%, P < 0.00001), and the results still showed a significant correlation between HSPs expression and OS, indicating that this study did not have a significant impact on the results.
Impact of HSPs on DFS, RFS, and CSS in CRC patients
Three articles reported the relationship between HSPs expression and DFS in CRC patients, with one article each reporting PFS and EFS. Given that the participants of the five articles were all CRC patients receiving conventional treatment, and the definitions of DFS, PFS, and EFS were essentially similar, a meta-analysis was conducted to examine the relationship between expression of HSPs and DFS in CRC patients. As shown in Fig. 4A, the overall HR for DFS in patients with high HSPs expression was 1.56 (95% CI: 1.18–2.06, I²=46%, P = 0.002). RFS data were extracted from two eligible studies, showing that the overall HR for RFS in patients with high HSPs expression was 2.15 (95% CI: 1.37–3.38, I²=0%, P = 0.0008, Fig. 4B). Two studies reported on CSS, with the overall HR for CSS being 1.65 (95% CI: 1.07–2.42, I²=0%, P = 0.0003, Fig. 4C). These results indicate that high expression of HSPs is significantly associated with poor DFS, RFS, and CSS, suggesting that high expression of HSPs may predict poor outcomes in CRC patients.
Fig. 4.
Forest plots for the association between the expression of HSPs and DFS (A), RFS (B), CSS (C) in CRC patients
Relationship between HSPs and clinicopathological characteristics in CRC patients
Data on patients with CRC degrees of differentiation, infiltration depth, lymph node metastasis, distant metastasis, TNM stage, tumor location, and tumor size, with high versus low expression of HSPs, were extracted from the literature, and a meta-analysis was performed to determine the relationship between high expression of HSPs and these clinicopathological characteristics. Tumors with T1-T2 grades were defined as having low infiltration, while those with T3-T4 grades were classified as high infiltration. Stage I-II tumors were considered low TNM stage, and stage III-IV tumors were considered high TNM stage. As shown in Table 3 and Supplementary Fig. S1, High expression of HSPs was significantly associated with higher TNM stage (OR = 1.29, 95% CI: 1.01–1.65, P = 0.04), but no significant correlation was found between high expression of HSPs and differentiation degree (OR = 0.92, 95% CI: 0.76–1.11, P = 0.39), infiltration depth (OR = 1.14, 95% CI: 0.93–1.40, P = 0.22), lymph node metastasis (OR = 1.06, 95% CI: 0.90–1.26, P = 0.48, distant metastasis (OR = 1.02, 95% CI: 0.56–1.86, P = 0.95,), tumor location (OR = 1.382, 95% CI: 1.01–1.88, P = 0.05), or tumor size (OR = 0.83, 95% CI: 0.62–1.11, P = 0.22).
Table 3.
HSPs expression in relation to clinicopathologic features of CRC patients
| Clinicopathological characteristics | No. of included cohorts | Effects model | Pooled OR with 95%CI | P | Heterogeneity | |
|---|---|---|---|---|---|---|
| I2 (%) | p | |||||
| differentiation degree | 11 | Fixed | 0.92 (0.76, 1.11) | 0.39 | 45 | 0.05 |
| infiltration depth | 10 | Fixed | 1.14 (0.93, 1.40) | 0.22 | 0 | 0.44 |
| lymph node metastasis | 10 | Fixed | 1.06 (0.9, 1.26) | 0.48 | 48 | 0.04 |
| distant metastasis | 5 | Random | 1.02 (0.56, 1.86) | 0.95 | 78 | 0.001 |
| TNM stage | 13 | Random | 1.29 (1.01, 1.65) | 0.04 | 57 | 0.005 |
| tumor location | 3 | Fixed | 1.382 (1.01, 1.88) | 0.05 | 0 | 0.46 |
| tumor size | 4 | Random | 0.83 (0.62, 1.11) | 0.22 | 52 | 0.10 |
Sensitivity analysis and publication bias
To ensure the stability of the meta-analysis, we conducted sensitivity analysis on 17 studies. By sequentially excluding each study and recalculating the pooled effect size, evaluate whether any individual studies have had an excessive impact on the overall results. Figure 5A demonstrates that the sensitivity analysis revealed no significant influence of any individual study on the overall HR, thereby confirming the stability and reliability of the results. Additionally, funnel plot, Begg’s test, and Egger’s test were used to assess publication bias. The funnel plot appeared visually symmetrical, and the symmetrical funnel plot suggested the absence of significant publication bias in the meta-analysis concerning OS (P = 0.174 for Begg’s test and P = 0.278 for Egger’s test, Fig. 5B). However, due to the limited number of studies (≤ 5) on the association between HSPs expression and DFS, RFS, and CSS, sensitivity analysis and publication bias analysis were not performed.
Fig. 5.
Sensitivity analysis (A) and funnel plot for publication bias (B) in terms of the association of the expression of HSPs with OS
Discussion
CRC is a common gastrointestinal tumor that seriously threatens people’s lives and health. In clinical practice, many patients may still experience recurrence and metastasis after surgery, radiotherapy, chemotherapy, or immunotherapy, and the overall prognosis is not ideal. Therefore, it is clinically important to actively understand the high-risk factors that affect the prognosis of patients with CRC. Some studies have demonstrated that the expression of HSPs may play a key role in tumor development,6 but the clinical evidence for their role in the prognosis of CRC patients is not yet uniformly conclusive. Although similar meta-analysis have been conducted in previous literature, but there is currently a ack of comprehensive meta-analysis integrating multiple HSPs subtypes to assess their impact on various survival outcomes (OS, DFS, etc.) and key clinicopathological characteristics. Most studies have only focused on a certain type of HSP and prognosis, such as Gao et al. [34] analyzing the relationship between HSP70 and OS prognosis in CRC patients. Our study conducted a comprehensive and integrated meta-analysis of multiple HSPs subtypes, and used multiple prognostic indicators such as OS, DFS, RFS, etc. as outcomes to further explored the prognostic value of HSPs expression in CRC patients and explore its detailed association with major clinical pathological features.
This meta-analysis included 3,919 CRC patients from different research centers, and the results showed that high expression of HSPs was significantly correlated with poor OS, DFS, RFS, and CSS, suggesting that high expression of HSPs has a significant adverse effect on the prognosis of CRC patients. This result is consistent with previous findings [35] and strongly supports the conclusion that high expression of HSPs may worsen the survival outcomes of CRC patients. Additionally, in the subgroup analysis of the impact of different types of HSPs on OS in CRC patients, we found that high expression of HSP70, HSP27, and HSP47 can predict poor OS, whereas HSP90 expression was not significantly correlated with OS in CRC patients. Gao et al. [34] conducted a meta-analysis on the role of HSP70 in the prognosis of CRC patients, reporting that HSP70 overexpression was associated with both OS and DFS in CRC patients. Therefore, it can be concluded that HSP70 overexpression predicts poor survival in CRC patients. Ge et al. [35] evaluated the prognostic value of HSPs in patients with gastrointestinal tumors but did not find any association between HSP90 and the prognosis of gastrointestinal tumor patients. However, there is currently limited literature on the relationship between HSP90 and the prognosis of CRC patients, and this study only included 2 articles, which may have limitations in statistical power. Therefore, this result requires further clinical research to explore. To further explore the role of HSPs in CRC, we also analyzed the relationship between the status of HSPs and the clinical pathological characteristics of CRC patients, which revealed that high expression of HSPs was significantly correlated with high TNM stage, a key factor in evaluating CRC prognosis. Patients with high TNM staging usually have a poorer prognosis, which also explains to a certain extent why high expression of HSPs adversely affects the prognosis of CRC patients. Moreover, this result helps refine risk stratification and assists clinicians in identifying patients who may benefit from more intensive adjuvant therapy or closer follow-up.
The heterogeneity of the overall OS meta-analysis results in this study was significant and difficult to explain from different HSPs subtypes, regions, sample sizes, sample types, and detection methods. However, we observed in each subgroup that CRC patients with high HSPs expression had poorer OS compared to those with low HSPs expression, further confirming the reliability of the conclusion. By removing one study at a time to examine heterogeneity, it was ultimately found that Banerjee et al. (2005) was the source of heterogeneity, which might be related to the fact that this study only included MSI-H CRC patients. The meta-analysis results after removing this study still showed a significant correlation between HSPs expression and OS, and sensitivity analysis did not show that any studies had a significant impact on the results. Therefore, the conclusion of this meta-analysis is relatively reliable.
HSPs are a type of protective protein synthesized under stress conditions, which can maintain the stability and function of intracellular proteins, resist various environmental and physiological pressures, and are important regulatory factors for cell proliferation and differentiation [36]. The proliferation and survival of tumor cells are more dependent on HSP chaperones than normal cells, as oncoproteins in cancer cells are frequently misfolded and require enhanced chaperone activity for correction [37]. An increasing number of studies have shown that different isoforms of HSPs are highly expressed in CRC and promote CRC progression with metastasis or drug resistance, which is detrimental to the prognosis of patients with CRC. This may be directly related to their structural and biological roles [38–40]. HSP70 plays an anti-apoptotic role in colorectal cancer by interacting with p53 to enhance its cytoplasmic segregation in CRC cell lines and inhibiting the transcriptional activation of p53 [41]. Moreover, silencing HSP70 triggers apoptosis in the colorectal cancer cell line HT29, effectively inhibiting the growth of tumors in colorectal cancer mice [42]. Additionally, high expression of HSP70 can interfere with the anti-tumor activity of proteasome inhibitors, leading to apoptosis resistance in the colon cancer cell line HCT-116 [43]. HSP27 has been shown to be overexpressed in colon cancer [38]. In CRC, HSP27 not only affects proliferation, migration, and invasion but also influences the cell cycle by promoting the transition from G2 phase to M phase and the epithelial-mesenchymal transition pathway. HSP27 can also play a role in the Ca2+ influx signaling pathway by maintaining the stability of STIM1 protein. Silencing HSP27 can inhibit the progression and metastasis of CRC by reducing STIM1 protein levels [44]. Chern et al. [40] reported that HSP47 supports the growth of CRC both in vitro and in vivo, confirming that HSP47 continuously activates AKT signaling by reducing the stability of PHLPP1, promotes the proliferation of colorectal cancer cell lines, and inhibits the efficacy of chemotherapy. This provides evidence that high expression of HSPs is associated with poor prognosis in patients with CRC.
Multiple studies have shown that HSP90 is highly expressed in various gastrointestinal tumors such as colorectal cancer [45]。In CRC, Tumor cells promote tumor protein stability through Hsp90 mediation, protecting it from degradation and thus promoting tumor occurrence and progression [46]. In the CRC mouse model, simultaneous inhibition of HSP90 and HSF1 can suppress tumor growth and reshape immune cell composition [47]. Moreover, clinical and basic research on HSP90 inhibitors is constantly being explored and has achieved good results [48–50]. The safety and anti-tumor activity of oral HSP90 inhibitor TAS-116 160 mg combined with nivolumab are controllable, especially suitable for MSS-CRC [50]. Although our research results did not show a significant relationship between HSP90 expression and the prognosis of CRC patients, this might be related to the limited number of included literature, and further clinical studies are needed to explore it.
Despite our efforts to improve this meta-analysis, there are still limitations. Firstly, the included studies were all retrospective and observational and their selection criteria were not as uniform as those in prospective randomized controlled trials, making it difficult to ensure the homogeneity of the research. Secondly, although we conducted a comprehensive search of multiple databases and both Begg’s and Egger’s tests showed no significant publication bias, this study only included literature published in English, which cannot completely eliminate publication bias. Thirdly, the heterogeneity of the meta-analysis results was significant, which may be related to the fact that the HR included in the study was the result of a multivariate analysis, with different confounding factors in each study. Additionally, we conducted a combined analysis of DFS, PFS, and EFS, which may have contributed to the heterogeneity of the study.
Conclusion
In conclusion, this meta-analysis shows that high expression of HSPs has a significant adverse effect on the prognosis of CRC patients. CRC patients with high expression of HSPs have shorter OS, DFS, RFS, and CSS, especially those with high expression of HSP70, HSP27, and HSP47, which can predict poor OS. In addition, high expression of HSPs is significantly correlated with high TNM stage. Based on current research, it is anticipated that HSPs, especially HSP70, HSP27, and HSP47, will emerge as novel molecular markers for the prognostic assessment of CRC patients. This feature helps refine risk stratification, assist clinicians in identifying individuals who may benefit from more intensive adjuvant therapy or closer follow-up, and aid in the development of non-invasive liquid biopsy markers for prognostic monitoring. However, prior to this occurrence, large-scale prospective cohort studies are necessary to furnish more credible evidence for the correlation between HSPs and the prognosis of CRC patients. In addition to verifying the prognostic role of HSPs, future research can also explore their predictive value for specific treatments such as chemotherapy or immunotherapy, and should focus on developing and validating standardized quantitative detection methods for HSPs panels. We believe that further research on the relationship between HSPs and the prognosis of CRC patients will provide strong evidence for clinical decision-making.
Supplementary Information
Acknowledgements
We thank all of our colleagues who contributed to this manuscript.
Abbreviations
- CI
Confidence interval
- CRC
Colorectal cancer
- CSS
Cancer-specific survival
- DFS
Disease-free survival
- EFS
Event-free survival
- HR
Hazard ratio
- HSP
Heat shock protein
- OS
Overall survival
- PFS
Progression-free survival
- RFS
Recurrence-free survival
Authors’ contributions
Z. W. (Conceptualization, Methodology, Data curation, Writing—original draft), J. Z. (Formal analysis, Visualization), and Z. Z. (Methodology, Data curation, Writing—original draft). All authors read and approved the final manuscript.
Funding
This work was supported by grants from the Cultivation Project (Nature) of Capital Medical University (PYZ24068) and Beijing Municipal Science and Technology Development Fund Project of Traditional Chinese Medicine (BJZYQN-2025-12).
Data availability
The data underlying this article will be shared on reasonable request by the corresponding author.
Declarations
Ethics approval and consent to participate
Not applicable.
Consent for publication
Not Applicable.
Competing interests
The authors declare no competing interests.
Footnotes
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Contributor Information
Yangfan Li, Email: lyfdzyx@vip.163.com.
Zhenmin Zhang, Email: doc_zzm@126.com.
References
- 1.Bray F, Laversanne M, Sung H, Ferlay J, Siegel RL, Soerjomataram I, et al. Global cancer statistics 2022: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries. CA Cancer J Clin. 2024;74(3):229–63. 10.3322/caac.21834. [DOI] [PubMed] [Google Scholar]
- 2.Rosati G, Aprile G, Colombo A, Cordio S, Giampaglia M, Cappetta A, et al. Colorectal cancer heterogeneity and the impact on precision medicine and therapy efficacy. Biomedicines. 2022;10(5):1035. 10.3390/biomedicines10051035. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Shinji S, Yamada T, Matsuda A, Sonoda H, Ohta R, Iwai T, et al. Recent advances in the treatment of colorectal cancer: A review. J Nippon Med Sch. 2022;89(3):246–54. 10.1272/jnms.JNMS.2022_89-310. [DOI] [PubMed] [Google Scholar]
- 4.Zuo WF, Pang Q, Zhu X, Yang QQ, Zhao Q, He G, et al. Heat shock proteins as hallmarks of cancer: insights from molecular mechanisms to therapeutic strategies. J Hematol Oncol. 2024;17(1):81. 10.1186/s13045-024-01601-1. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Youness RA, Gohar A, Kiriacos CJ, El-Shazly M. Heat shock proteins: central players in oncological and Immuno-Oncological tracks. Adv Exp Med Biol. 2023;1409:193–203. 10.1007/5584_2022_736. [DOI] [PubMed] [Google Scholar]
- 6.Mittal S, Rajala MS. Heat shock proteins as biomarkers of lung cancer. Cancer Biol Ther. 2020;21(6):477–85. 10.1080/15384047.2020.1736482. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Safi S, Messner L, Kliebisch M, Eggert L, Ceylangil C, Lennartz P, et al. Circulating Hsp70 levels and the immunophenotype of peripheral blood lymphocytes as potential biomarkers for advanced lung cancer and therapy failure after surgery. Biomolecules. 2023;13(5):874. 10.3390/biom13050874. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Hrudka J, Jelínková K, Fišerová H, Matěj R, Mandys V, Waldauf P. Heat shock proteins 27, 70, and 110: expression and prognostic significance in colorectal cancer. Cancers (Basel). 2021;13(17):4407. 10.3390/cancers13174407. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Tustumi F, Agareno GA, Galletti RP, da Silva RBR, Quintas JG, Sesconetto LA, et al. The role of the Heat-Shock proteins in esophagogastric cancer. Cells. 2022;11(17):2664. 10.3390/cells11172664. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Zhang M, Bi X. Heat shock proteins and breast cancer. Int J Mol Sci. 2024;25(2):876. 10.3390/ijms25020876. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Sun C, Pan Q, Du M, Zheng J, Bai M, Sun W. Decoding the roles of heat shock proteins in liver cancer. Cytokine Growth Factor Rev. 2024;75:81–92. 10.1016/j.cytogfr.2023.12.003. [DOI] [PubMed] [Google Scholar]
- 12.Drecoll E, Nitsche U, Bauer K, Berezowska S, Slotta-Huspenina J, Rosenberg R. Expression analysis of heat shock protein 90 (HSP90) and Her2 in colon carcinoma. Int J Colorectal Dis. 2014;29(6):663–71. 10.1007/s00384-014-1857-3. [DOI] [PubMed] [Google Scholar]
- 13.Bauer K, Nitsche U, Slotta-Huspenina J, Drecoll E, von Weyhern CH, Rosenberg R, et al. High HSP27 and HSP70 expression levels are independent adverse prognostic factors in primary resected colon cancer. Cell Oncol (Dordr). 2012;35(3):197–205. 10.1007/s13402-012-0079-3. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Page MJ, McKenzie JE, Bossuyt PM, Boutron I, Hoffmann TC, Mulrow CD, et al. The PRISMA 2020 statement: an updated guideline for reporting systematic reviews. BMJ. 2021;372:n71. 10.1136/bmj.n71. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Stang A. Critical evaluation of the Newcastle-Ottawa scale for the assessment of the quality of onrandomized studies in meta-analyses. Eur J Epidemiol. 2010;25:603–605. https://doi.org/0.1007/s10654-010-9491-z. [DOI] [PubMed]
- 16.Banerjea A, Feakins RM, Nickols CD, Phillips SM, Powar MP, Bustin SA, et al. Immunogenic hsp-70 is overexpressed in colorectal cancers with high-degree microsatellite instability. Dis Colon Rectum. 2005;48(12):2322–8. 10.1007/s10350-005-0203-9. [DOI] [PubMed] [Google Scholar]
- 17.Gráf L, Barabás L, Madaras B, Garam N, Maláti É, Horváth L, et al. High serum Hsp70 level predicts poor survival in colorectal cancer: results obtained in an independent validation cohort. Cancer Biomark. 2018;23(4):539–47. 10.3233/CBM-181683. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Jubran R, Kocsis J, Garam N, Maláti É, Gombos T, Barabás L, et al. Circulating mitochondrial stress 70 protein/mortalin and cytosolic Hsp70 in blood: risk indicators in colorectal cancer. Int J Cancer. 2017;141(11):2329–35. 10.1002/ijc.30918. [DOI] [PubMed] [Google Scholar]
- 19.Kalioraki MA, Artemaki PI, Sklirou AD, Kontos CK, Adamopoulos PG, Papadopoulos IN, et al. Heat shock protein beta 3 (HSPB3) is an unfavorable molecular biomarker in colorectal adenocarcinoma. Mol Carcinog. 2020;59(1):116–25. 10.1002/mc.23133. [DOI] [PubMed] [Google Scholar]
- 20.Kang M, Jeong S, An J, Park S, Nam S, Kwon KA, et al. Clinicopathologic significance of heat shock protein 60 as a survival predictor in colorectal cancer. Cancers (Basel). 2023;15(16):4052. 10.3390/cancers15164052. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Kim JH, Kim KJ, Rhee YY, Oh S, Cho NY, Lee HS, et al. Expression status of wild-type HSP110 correlates with HSP110 T17 deletion size and patient prognosis in microsatellite-unstable colorectal cancer. Mod Pathol. 2014;27(3):443–53. 10.1038/modpathol.2013.160. [DOI] [PubMed] [Google Scholar]
- 22.Kocsis J, Madaras B, Tóth EK, Füst G, Prohászka Z. Serum level of soluble 70-kD heat shock protein is associated with high mortality in patients with colorectal cancer without distant metastasis. Cell Stress Chaperones. 2010;15(2):143–51. 10.1007/s12192-009-0128-7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Lazaris AC, Theodoropoulos GE, Davaris PS, Panoussopoulos D, Nakopoulou L, Kittas C, et al. Heat shock protein 70 and HLA-DR molecules tissue expression. Prognostic implications in colorectal cancer. Dis Colon Rectum. 1995;38(7):739–45. 10.1007/BF02048033. [DOI] [PubMed] [Google Scholar]
- 24.Mori K, Toiyama Y, Okugawa Y, Ichikawa T, Nagano Y, Oki S, et al. Preoperative heat shock protein 47 levels identify colorectal cancer patients with lymph node metastasis and poor prognosis. Oncol Lett. 2020;20(6):333. 10.3892/ol.2020.12196. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Mori K, Toiyama Y, Otake K, Fujikawa H, Saigusa S, Hiro J, et al. Proteomics analysis of differential protein expression identifies heat shock protein 47 as a predictive marker for lymph node metastasis in patients with colorectal cancer. Int J Cancer. 2017;140(6):1425–35. 10.1002/ijc.30557. [DOI] [PubMed] [Google Scholar]
- 26.Oh HJ, Kim JH, Lee TH, Park HE, Bae JM, Lee HS, et al. Dominant high expression of wild-type HSP110 defines a poor prognostic subgroup of colorectal carcinomas with microsatellite instability: a whole-section immunohistochemical analysis. APMIS. 2017;125(12):1076–83. 10.1111/apm.12770. [DOI] [PubMed] [Google Scholar]
- 27.Rozenberg P, Kocsis J, Saar M, Prohászka Z, Füst G, Fishelson Z. Elevated levels of mitochondrial Mortalin and cytosolic HSP70 in blood as risk factors in patients with colorectal cancer. Int J Cancer. 2013;133(2):514–8. 10.1002/ijc.28029. [DOI] [PubMed] [Google Scholar]
- 28.Schweiger T, Nikolowsky C, Starlinger P, Traxler D, Zimmermann M, Birner P, et al. Stromal expression of heat-shock protein 27 is associated with worse clinical outcome in patients with colorectal cancer lung metastases. PLoS ONE. 2015;10(3):e0120724. 10.1371/journal.pone.0120724. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Tweedle EM, Khattak I, Ang CW, Nedjadi T, Jenkins R, Park BK, et al. Low molecular weight heat shock protein HSP27 is a prognostic indicator in rectal cancer but not colon cancer. Gut. 2010;59(11):1501–10. 10.1136/gut.2009.196626. [DOI] [PubMed] [Google Scholar]
- 30.Wang F, Zhang P, Shi C, Yang Y, Qin H. Immunohistochemical detection of HSP27 and HnRNP K as prognostic and predictive biomarkers for colorectal cancer. Med Oncol. 2012;29(3):1780–8. 10.1007/s12032-011-0037-3. [DOI] [PubMed] [Google Scholar]
- 31.Yu Z, Zhi J, Peng X, Zhong X, Xu A. Clinical significance of HSP27 expression in colorectal cancer. Mol Med Rep. 2010;3(6):953–8. 10.3892/mmr.2010.372. [DOI] [PubMed] [Google Scholar]
- 32.Zhang Y, Ni L, Li Q, Li M, Zhu J, Zhang F. Diagnostic, clinicopathologic, therapeutic and prognostic value of plasma heat shock protein 90 levels in patients with advanced Gastrointestinal carcinoma. J Cancer. 2020;11(20):5890–9. 10.7150/jca.46343. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Zhang S, Guo S, Li Z, Li D, Zhan Q. High expression of HSP90 is associated with poor prognosis in patients with colorectal cancer. PeerJ. 2019;7:e7946. 10.7717/peerj.7946. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Gao G, Liu S, Yao Z, Zhan Y, Chen W, Liu Y. The prognostic significance of Hsp70 in patients with colorectal cancer patients: A PRISMA-Compliant Meta-Analysis. Biomed Res Int. 2021;2021:5526327. 10.1155/2021/5526327. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35.Ge H, Yan Y, Guo L, Tian F, Wu D. Prognostic role of HSPs in human Gastrointestinal cancer: a systematic review and meta-analysis. Onco Targets Ther. 2018;11:351–9. 10.2147/OTT.S155816. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36.Mazurakova A, Solarova Z, Koklesova L, Caprnda M, Prosecky R, Khakymov A. Heat shock proteins in cancer - Known but always being rediscovered: their perspectives in cancer immunotherapy. Adv Med Sci. 2023;68(2):464–73. 10.1016/j.advms.2023.10.005. [DOI] [PubMed] [Google Scholar]
- 37.Chatterjee S, Burns TF. Targeting heat shock proteins in cancer: A promising therapeutic approach. Int J Mol Sci. 2017;18(9):1978. 10.3390/ijms18091978. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38.Asgharzadeh F, Moradi-Marjaneh R, Marjaneh MM. The role of heat shock protein 27 in carcinogenesis and treatment of colorectal cancer. Curr Pharm Des. 2022;28(32):2677–85. 10.2174/1381612828666220427140640. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 39.Soleimani A, Zahiri E, Ehtiati S, Norouzi M, Rahmani F, Fiuji H, et al. Therapeutic potency of heat-shock protein-70 in the pathogenesis of colorectal cancer: current status and perspectives. Biochem Cell Biol. 2019;97(2):85–90. 10.1139/bcb-2018-0177. [DOI] [PubMed] [Google Scholar]
- 40.Chern Y, Zhang P, Ju H, Tai IT. Heat shock protein 47 promotes tumor survival and therapy resistance by modulating AKT signaling via PHLPP1 in colorectal cancer. Cancer Biol Med. 2020;17(2):343–56. 10.20892/j.issn.2095-3941.2019.0261. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 41.Gestl EE, Anne Böttger S. Cytoplasmic sequestration of the tumor suppressor p53 by a heat shock protein 70 family member, mortalin, in human colorectal adenocarcinoma cell lines. Biochem Biophys Res Commun. 2012;423(2):411–6. 10.1016/j.bbrc.2012.05.139. [DOI] [PubMed] [Google Scholar]
- 42.Zhong MA, Zhang H, Qi XY, Lu AG, You TG, Gao W, et al. ShRNA-mediated gene Silencing of heat shock protein 70 inhibits human colon cancer growth. Mol Med Rep. 2011;4(5):805–10. 10.3892/mmr.2011.528. [DOI] [PubMed] [Google Scholar]
- 43.Ramapathiran L, Bernas T, Walter F, Williams L, Düssmann H, Concannon CG, et al. Single-cell imaging of the heat-shock response in colon cancer cells suggests that magnitude and length rather than time of onset determines resistance to apoptosis. J Cell Sci. 2014;127(Pt 3):609–19. 10.1242/jcs.137158. [DOI] [PubMed] [Google Scholar]
- 44.Huang CY, Wei PL, Chen WY, Chang WC, Chang YJ. Silencing heat shock protein 27 inhibits the progression and metastasis of colorectal cancer (CRC) by maintaining the stability of stromal interaction molecule 1 (STIM1) proteins. Cells. 2018;7(12):262. 10.3390/cells7120262. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 45.Liu L, Deng Y, Zheng Z, Deng Z, Zhang J, Li J, et al. Hsp90 inhibitor STA9090 sensitizes hepatocellular carcinoma to Hyperthermia-Induced DNA damage by suppressing DNA-PKcs protein stability and mRNA transcription. Mol Cancer Ther. 2021;20(10):1880–92. 10.1158/1535-7163. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 46.Klemke L, De Oliveira T, Witt D, Winkler N, Bohnenberger H, Bucala R, et al. Hsp90-stabilized MIF supports tumor progression via macrophage recruitment and angiogenesis in colorectal cancer. Cell Death Dis. 2021;12(2):155. 10.1038/s41419-021-03426-z. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 47.Isermann T, Schneider KL, Wegwitz F, De Oliveira T, Conradi LC, Volk V, et al. Enhancement of colorectal cancer therapy through interruption of the HSF1-HSP90 axis by p53 activation or cell cycle Inhibition. Cell Death Differ. 2025;32(9):1734–49. 10.1038/s41418-025-01502-x. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 48.Jia L, Yang H, Liu Y, Zhou Y, Li G, Zhou Q, et al. Targeted delivery of HSP90 inhibitors for efficient therapy of CD44-positive acute myeloid leukemia and solid tumor-colon cancer. J Nanobiotechnol. 2024;22(1):198. 10.1186/s12951-024-02460-1. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 49.Kryeziu K, Bruun J, Guren TK, Sveen A, Lothe RA. Combination therapies with HSP90 inhibitors against colorectal cancer. Biochim Biophys Acta Rev Cancer. 2019;1871(2):240–7. 10.1016/j.bbcan.2019.01.002. [DOI] [PubMed] [Google Scholar]
- 50.Kawazoe A, Itahashi K, Yamamoto N, Kotani D, Kuboki Y, Taniguchi H, et al. TAS-116 (Pimitespib), an oral HSP90 Inhibitor, in combination with nivolumab in patients with colorectal cancer and other solid tumors: an Open-Label, Dose-Finding, and expansion phase Ib trial (EPOC1704). Clin Cancer Res. 2021;27(24):6709–15. 10.1158/1078-0432. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
The data underlying this article will be shared on reasonable request by the corresponding author.