Key points are not available for this paper at this time.
Polo-like kinase 1 (PLK1), encoded by the PLK1 gene and mainly consisting of a Pkinase domain and two Polo-box domains (PB1 and PB2),1 is generally considered as a cancer promoter, owing to its critical role in cell cycle and overexpression.2 Recent studies have revealed that PLK1 might play as a tumour-suppressor gene,3, 4 and limited preclinical achievements of PLK1 inhibitors have been translated into good clinical outcomes.5 Considering concurrent driver and passenger mutations involved in developing treatment plan, genomic profiles are critical to PLK1-mutated patients receiving PLK1 inhibitor combination therapy. Herein, of 398 PLK1-mutated pan-cancer tissue samples including lung cancer, colorectal cancer (CRC), breast cancer, hepatobiliary cancer, genitourinary cancer, uterine cancer, etc., we enrolled 126 patients with lung cancer (the LC cohort) and 160 patients with CRC (the CRC cohort) to profile the genomic landscape of PLK1 and concomitant mutations under natural selection within lung cancer and CRC, detailed in Table S1 and Table S2, respectively (Figure 1A). In addition to comparing the genomic features of the CRC cohort to 160 PLK1 wild-type CRCs, further investigation between 63 PLK1-mutated (the cBioPortal CRC cohort) and wild-type CRCs in the cBioPortal database was performed to study the role of PLK1 in CRC development (See Appendix S1 for methods). The mutational landscapes and clinical characteristics of the CRC and LC cohorts were summarized in Figure S1A. The CRC cohort included 45 microsatellite instable (MSI) CRCs, 49 microsatellite stable (MSS) CRCs and 66 CRCs with unknown MSI status. TP53, APC, KRAS, KMT2B and PIK3CA were frequently mutated in CRCs harbouring PLK1 mutations. The LC cohort consisted of 105 patients with non-small cell lung cancer (NSCLC), three patients with small cell lung cancer and 18 lung cancer patients with unknown histological subtypes. Similar to the CRC cohort, TP53, KRAS and EGFR mutations were relatively common in the LC cohort, whereas APC, KMT2B and PIK3CA were less frequently altered. PLK1 mutation subtypes in the LC, CRC and cBioPortal CRC cohorts were summarized in Figure 1B. According to the genomic region, PLK1 mutations were categorized into the Pkinase (blue), PB1 (red), PB2 (brown) or other (grey) subgroup. Of note, neither genomic aberrations leading to amino acid substitutions at the position 131 of PLK1 (PLK1E131X) nor those at position 133 (PLK1C133X), which were related to the direct interaction between PLK1 inhibitors and the Pkinase domain,1 were detected in three study cohorts. Besides considerable PLK1 missense mutations, various truncating mutations were observed, especially in CRC, in which truncating mutations were more common than in lung cancer (p = 0.02, Figure S1B). Moreover, the prevalence of PLK1 truncating mutations was comparable between MSI and MSS CRCs (p = 0.89), which was higher than that of NSCLC (p = 0.04, Figure 1B). However, we were unable to identify obvious PLK1 hotspots under natural selection in the CRC or LC cohorts, which is non-intuitive if PLK1 plays as an oncogene. In the CRC cohort, compared to CRC harbouring PB1/2 mutations, PLK1 truncating mutations appeared to be more prevalent in the Pkinase domains or the regions other than Pkinase and PB1/2 (Other) (p = 0.05, Figure 1C), with no obvious differences between MSI and MSS CRCs (Figure S1C). By contrast, this trend could not be detected in NSCLCs of the LC cohort. PLK1-mutated CRCs exhibited higher tumour mutation burden (TMB) than PLK1-mutated NSCLCs, which were the majority of the LC cohort (Figure S1D). In the CRC cohort, TMB appeared to be higher in patients harbouring PLK1 Other mutations than in those with PLK1 Pkinase (p = 0.04), PB1 (p < 0.001) or PB2 (p = 0.04) mutations (Figure 1D). Intriguingly, the difference in prevalence of POLE somatic mutations was similar to the TMB levels across four subgroups (Figure 1D). The results of a multivariable linear regression model adjusting for MSI and POLE mutational status demonstrated relatively high TMB in the Other subgroup (vs. PB1, p = 0.04; vs. PB2, p = 0.06) and no significant TMB difference between MSI and MSS CRCs (p = 0.45, Figure 1D). Additionally, PLK1 Other mutations might be associated with higher prevalence of mutated cell cycle pathway in both NSCLC and CRC, particularly in comparison to PLK1 PB1 mutations (p < 0.001, Figure 1E). Consistent with our findings in lung cancer TMB, there were no mutations obviously enriched in lung cancers with PLK1 Other mutations (Figure S1E), whereas considerable mutated genes were more common in CRCs with PLK1 Other mutations (Figure 1F). Interestingly, mutations leading to amino acid substitutions at the position 12 of KRAS (KRASG12X) were more prevalent in CRCs carrying PLK1 PB1 mutations (p < 0.001, Figure 1G), suggesting a potential cooccurrence of KRAS dysfunction and PLK1 PB1 domain abnormalities in CRC. PLK1-mutated CRCs with concurrent KRAS mutations had higher PLK1 mutation variant allele frequencies than those without (p = 0.03, Figure 1H). Additionally, TP53 mutations exhibiting cooccurrence with PLK1 mutations in Pkinase, PB1 or PB2 regions were exclusively observed in MSS CRCs while multiple genetic alterations were enriched in MSS CRCs with PLK1 Other mutations (Figure 1I). As our findings suggested a stronger association of PLK1 mutations with concomitant mutations in CRC than in lung cancer, we further performed comparison of other genomic features between PLK1-mutated and wild-type CRCs. In the cBioPortal database, MSI was more prevalent in PLK1-mutated CRCs than in those with wild-type counterparts (p < 0.001, Figure 2A). Compared to PLK1 wild-type CRCs, the CRC cohort showed higher TMB (p < 0.001, Figure 2B) whereas similar TMB levels were observed in CRCs with PLK1 truncating and non-truncating mutations (p = 0.34), suggesting that more genetic abnormalities were accumulated in PLK1-mutated CRCs without obvious differences between truncating and non-truncating mutations. Intriguingly, the prevalence of APC and KRAS mutations, which are well-known CRC drivers, was comparable between CRCs with and without PLK1 mutations; however, TP53 mutations were more prevalent in PLK1 wild-type CRCs while considerable mutated genes were enriched in PLK1-mutated CRCs (Figure 2C). Moreover, TP53 mutations were common in CRCs with non-truncating PLK1 Other mutations than those with truncating PLK1 Other mutations (p = 0.01, Figure 2D). We supposed that non-truncating PLK1 mutations in non-functional domains might have limited influence on CRC development, requiring more involvement of TP53 dysfunction. Furthermore, relatively high chromosome instability scores were observed in PLK1 wild-type CRCs in comparison to CRCs harbouring PLK1 non-truncating mutations (p = 0.07, Figure 2E) when the confounding effect of MSI (p < 0.001) and POLE (p = 0.03) mutational status were controlled for; however, the difference between CRCs with PLK1 truncating mutations and with wild-type PLK1 was no longer significant (p = 0.37). To summarize, based on the genomic landscape under natural selection within lung cancer and CRC tumours, the clinical application of PLK1 inhibitors still needs to be approached with caution given potent tumour-suppressor features of PLK1. Our study had some limitations that need to be considered. First, most patients enrolled in this study had missing information about cancer stage, treatment and prognosis data, which restricted the study of the association between PLK1 mutation subtypes and treatment efficacy. Larger PLK1-mutated CRC and lung cancer populations are also required to explore the interaction between PLK1 and KRAS mutations in CRC and compare the molecular features between PLK1-mutated NSCLC and small cell lung cancer, respectively. In addition, our findings were mainly based on CRC and NSCLC samples, and further studies focused on other cancer types are needed. Finally, future experimental investigation and validation are warranted to functionally determine whether PLK1 mutations would affect PLK1 inhibitor binding. Shuo Wang: Data curation (lead); formal analysis (equal); funding acquisition (equal); investigation (equal); validation (equal); writing – original draft (lead); writing – review and editing (equal). Feng Gao: Data curation (lead); formal analysis (equal); investigation (equal); resources (equal); validation (equal); writing – original draft (lead); writing – review and editing (equal). Yinghui Bi: Data curation (equal); formal analysis (equal); investigation (equal); writing – original draft (equal); writing – review and editing (equal). Xiaotian Zhao: Formal analysis (equal); investigation (equal); methodology (equal); software (equal); visualization (equal); writing – original draft (equal); writing – review and editing (equal). Qiuxiang Ou: Formal analysis (equal); investigation (equal); project administration (supporting); visualization (equal); writing – original draft (equal); writing – review and editing (equal). Minyi Zhu: Formal analysis (equal); investigation (equal); visualization (equal); writing – original draft (equal); writing – review and editing (equal). Xue Wu: Formal analysis (equal); investigation (equal); project administration (supporting); supervision (supporting); writing – original draft (equal); writing – review and editing (equal). Xuefei Zhang: Conceptualization (equal); funding acquisition (equal); investigation (supporting); project administration (equal); resources (equal); supervision (lead); writing – original draft (equal); writing – review and editing (equal). Kaiping Mao: Conceptualization (equal); methodology (equal); project administration (equal); resources (equal); supervision (lead); writing – original draft (equal); writing – review and editing (equal). The authors thank all the patients who participated in this study. This study was supported by the Science foundation of Peking University Cancer Hospital (No. 2021-7, to Shuo Wang), the Special Fund for Clinical Research of Wu Jieping Medical Foundation (No. 320.6750.2021-16-8, to Xuefei Zhang), and '1 + X' program for Clinical Competency Enhancement-Clinical Research Incubation Project, the Second Hospital of Dalian Medical University (No. 2022LCYJYB09, to Xuefei Zhang). Xiaotian Zhao, Qiuxiang Ou, Minyi Zhu, and Xue Wu are employees of Nanjing Geneseeq Technology Inc., China. The remaining authors have nothing to disclose. The mutation list of 286 tissue samples is provided as the Table S2, and other datasets used and/or analysed during the current study are available from the corresponding author on reasonable request. The external combined CRC dataset was obtained from the cBioPortal database (https://bit.ly/43D5zsm). Figure S1. Table S1. Table S2. Appendix S1. Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.
Wang et al. (Sat,) studied this question.