Genetic variability of HIF1A and response to treatment with cisplatin in combination with pemetrexed or gemcitabine in patients with malignant mesothelioma

Malignant mesothelioma (MM) is a rare and highly aggressive cancer that most commonly arises from the mesothelial cells of the pleura.^1,2 In vast majority of cases, it is associated with direct occupational or environmental exposure to asbestos.^3,4 MM has a long latency period, and its incidence continues to rise both in Slovenia and worldwide.^3,5,6 It typically presents with dyspnoea, chest pain, weight loss, and fatigue. Most patients are already in an advanced stage of the disease at the time these symptoms appear.^7,8

The cornerstone of treatment for MM remains systemic chemotherapy with cisplatin, mostly combined either with pemetrexed or gemcitabine. Chemotherapy is part of a multimodal treatment approach for patients with operable MM, administered either before or after surgery. Combination chemotherapy has proven to be more effective than monotherapy. Additionally, chemotherapy is used for palliative treatment in patients with inoperable disease, advanced stages, or those who decline surgery. In the 1990s, a trimodal treatment approach became established for operable MM, which includes adjuvant or neoadjuvant chemotherapy, surgery, and adjuvant radiotherapy.^9–12 Systemic treatment combining chemotherapy and immune checkpoint inhibitors (ICIs) is often chosen for patients with inoperable or recurrent MM who are in good general condition. Early inclusion of patients receiving systemic treatment for inoperable MM in palliative care is important, as it has been shown to prolong survival.^11,13–15 Additionally, a newer treatment option for inoperable MM is immunotherapy using ICIs such as nivolumab plus ipilimumab, which can be used as first-line therapy or introduced in the second line of treatment.^12,16–18

Despite advances in chemotherapy and immunotherapy, MM remains largely incurable, prompting ongoing research into biological markers for more personalized treatment.^12,19,20 Hypoxia was shown to reduce the responsiveness of MM cells to cisplatin while increasing their invasiveness and the synthesis of hypoxia-inducible factor 1 (HIF-1), suggesting that HIF-1 could represent a significant target for potential new therapeutic approaches in MM.²¹

The predominance of tumor cell proliferation over angiogenesis, along with the abnormal structure and distribution of the capillary network in MM tissue, creates hypoxic conditions to which tumor cells adapt in various ways. The primary homeostatic response to hypoxia is mediated by transcription factors from the hypoxia-inducible factor (HIF) family. These intracellular proteins become activated under hypoxic conditions, bind to DNA, and regulate gene expression, enabling cells to adapt to oxygen deprivation. HIFs exist in three different isoforms, with HIF-1 expressed in all cells, while HIF-2 and HIF-3 are present only in certain tissues. Expression of HIF-1 in tumors increases the transcription of genes encoding enzymes, growth factors, and other proteins, significantly influencing tumor growth, survival, angiogenesis, and metastasis.^22–24

HIF-1 is a heterodimeric protein composed of an oxygen-regulated alpha subunit (HIF-1A) and an oxygen-insensitive, constitutively expressed beta subunit (HIF-1B). It is tightly regulated to ensure its effects occur only under hypoxic conditions.^22,25 Genetic variants may impair HIF-1 regulation, resulting in HIF-1 being active even under normoxic conditions, leading to autonomous heterodimerization of the HIF-1 subunits, DNA binding, and transcription of target genes independent of oxygen levels in the cells.²³

Genetic variants, most commonly single nucleotide polymorphisms (SNPs) in the gene encoding the alpha subunit of HIF-1 (HIF1A), were reported to be associated with various diseases, with notable links to different malignancies. The most frequently studied SNPs are rs11549465 and rs11549467. In SNP rs11549465, cytosine (C) at position 1772 is replaced by thymine (T), resulting in the substitution of proline 582 with serine (p.Pro582Ser) within the oxygen-dependent degradation domain (ODD). In SNP rs11549467, adenine (A) at position 1790 is replaced by guanine (G), leading to the replacement of alanine 588 with threonine (p.Ala588Thr) in the ODD.²³ Another SNP often linked to cancer is rs2057482 (c.A395G), which involves an A to G substitution in the 3’ untranslated region (3’ UTR) of the gene.²⁶

A review study that analyzed 97 association studies identified links between 16 HIF1A SNPs and 40 disease phenotypes, including 14 types of cancer, such as breast cancer, lung cancer, colorectal cancer, stomach cancer, prostate cancer, oral cavity cancer, and renal cell carcinoma. Among the most frequently reported cancer-associated SNPs are HIF1A rs11549465, rs11549467, and rs2057482.²³ Additionally, studies have found that HIF1A SNPs are associated with worse disease prognosis of cancer patients and an increased risk of metastasis.^23,27,28 Most studies conducted to date have focused on the association of HIF-1A with cancer risk, but only a few investigated treatment response and survival. To date, however, no study has examined the role of HIF-1A in MM.

The aim of this study was to investigate the role of HIF-1A in the response to chemotherapy in patients with MM. We hypothesized that HIF1A polymorphisms are associated with the response to treatment with a combination of cisplatin and pemetrexed or gemcitabine in patients with MM.

We conducted a retrospective genetic association study. Our study population consisted of 234 patients with MM who received treatment with cisplatin combined with either pemetrexed or gemcitabine between January 1st, 2001, and September 30, 2018, at the Institute of Oncology Ljubljana, Slovenia. Diagnostic tissue samples from patients with pleural MM were collected using medical thoracoscopy or video-assisted thoracoscopic surgery (VATS), while samples for peritoneal MM were obtained using laparoscopic or open surgical procedures. Confirmation of diagnosis was performed through histopathological evaluation supported by immunohistochemical staining. Comprehensive clinical data, including patient demographics, therapeutic regimens, treatment responses, survival outcomes and adverse events, were extracted from patients’ medical records of the Institute of Oncology Ljubljana and the Cancer Registry of the Republic of Slovenia. Treatment response was categorized into complete response (CR), partial response (PR), stable disease (SD), and progressive disease (PD).

This study is the first to investigate the association of HIF1A SNPs with response to treatment with cisplatin in combination with pemetrexed or gemcitabine, as well as with PFS and OS in patients with MM. The key finding was a significantly poorer response to chemotherapy in patients carrying the rs11549465 CT genotype compared to CC genotype carriers, after adjusting for weight loss and CRP. No associations were found for rs11549467 or rs2057482 with treatment response in MM. Furthermore, none of the investigated SNPs were associated with PFS or OS.

We are the first to report a statistically significant poorer response to chemotherapy in patients with MM associated with the rs11549465 CT genotype. No other study has investigated the association between HIF1A rs11549465 and response to chemotherapy and survival in MM, and there were only a few studies in other cancers. Furthermore, most of these studies were conducted by Chinese researchers and are thus limited to Asian populations.²³ A study in 741 Chinese patients with histologically confirmed stage III or IV non-small cell lung cancer (NSCLC) reported that carriers of the rs11549465 CC genotype were more likely to respond to platinum-based chemotherapy than those with the TT genotype. They also reported that among NSCLC patients, a higher proportion of rs11549465 TT genotype carriers had higher HIF-1A expression compared to those with the CC genotype, and patients with high HIF-1A expression were significantly more likely to have a poor response to chemotherapy when compared to patients with low expression.³⁰ While we observed no association of rs11549465 with OS and PFS in MM patients, among NSCLS patients longer OS and PFS was reported for CC genotype carriers compared to CT or TT genotype carriers.³⁰ Another study examining the association between this polymorphisms and response to platinum-based neoadjuvant chemotherapy in locally advanced cervical cancer found a higher likelihood of treatment response in carriers of the CC genotype compared to those with CT or TT genotypes.³¹ Their results are consistent with our finding in MM that the CT genotype is associated with a poorer chemotherapy response. Increased HIF-1A expression and poorer disease prognosis in rs11549465 CT or TT carriers were also reported in a recent study on advanced NSCLC treated with radiotherapy.³²

A poorer response to chemotherapy in patients carrying the CT genotype could be partially explained by its association with increased expression of HIF-1A. Studies have shown that higher HIF-1A expression activates genes involved in pathways regulating cell proliferation, drug metabolism, apoptosis, autophagy, and angiogenesis, all of which contribute to treatment resistance. The resulting dysregulation may also affect the genes encoding drug transporters such as multidrug resistance protein 1 (MDR1) and multidrug resistance-associated protein 2 (MRP2), the glucose transporter 1 (GLUT-1), the vascular endothelial growth factor (VEGF), and the anti-apoptotic protein Bcl-2.^33,34 It has a been also shown in isogenic fibrosarcoma cells, that inactivation of HIF-1A can reverse hypoxia-induced cisplatin resistance.³⁵ However, this observations do not fully align with our findings, since increased HIF-1A expression was also associated with the rs11549465 TT genotype, not associated with chemotherapy response in our study. Such discrepancies between our results and those of related studies may be due to the complex and diverse mechanisms through which HIF-1A promotes chemo- and radio-resistance in cancer cells.³⁴

We found no association between rs11549467 and chemotherapy response, PFS and OS in patients with MM undergoing platinum-based chemotherapy. To our knowledge, no previous studies have investigated this polymorphism in MM patients, while the aforementioned studies in NSCLC and cervical cancer similarly reported no association between rs11549467 and chemotherapy response or survival.^30–32

In our study, rs2057482 was also not associated with chemotherapy response, PFS or OS in MM patients treated with chemotherapy. To our knowledge, no studies have investigated this polymorphism in MM patients. However, in some other cancers, the findings regarding this polymorphism have been quite contradictory and cannot be directly compared to our results due to different treatment. A study examining the association of HIF1A polymorphisms with prognosis in patients with advanced NSCLC treated with radiotherapy found that rs2057482 CT or TT genotypes carriers had shorter PFS and worse OS compared to the CC genotype carriers.³² On the other hand, in patients with early-stage (TNM I/II) NSCLC undergoing surgical treatment, the CT and TT genotypes seemed to have a protective effect on OS and DFS compared to the CC genotype. However, when patients of all stages were included in the analysis, no differences were observed between genotypes.³⁶ Similarly, in aggressive hepatocellular carcinoma, longer DFS and borderline longer OS following surgical treatment were observed in of the CT or TT genotypes carriers.³⁷

These discrepancies suggest that the impact of HIF1A polymorphisms on patient survival may be cancer-type or treatment specific. A genotype that is protective in one cancer type could be associated with poorer prognosis in another.³² This may be explained by the fact that rs2057482 is located in the 3’ UTR region of the HIF1A, near two microRNA binding sites. Studies have shown that polymorphisms in this region can alter microRNA binding sites, thereby affecting mRNA stability. This, in turn, could modify HIF1A expression and consequently influence survival outcomes.^37,38 Further studies are needed to confirm these hypotheses.

A potential limitation of our study may be, that we observed a deviation of rs11549465 polymorphism from HWE. To verify the accuracy of the KASP genotyping results, we repeated genotyping of this polymorphism in all samples using an independent TaqMan assay and the results were concordant. Despite this, the genotype distribution remained inconsistent with HWE, possibly due to higher frequency of homozygotes for the rare allele (TT). This observed deviation from HWE for rs11549465 may also reflect the nature of our patient cohort, which does not represent a general population as MM is a rare cancer, mostly associated with asbestos exposure. Moreover, studies have shown that HIF1A polymorphisms are associated with the risk of developing various cancers.^13,39,40

The strength of our study lies in the inclusion of as many as 234 patients with MM, which, considering the rarity of this disease, represents a very large cohort and one of the largest genetic association studies in MM both nationally and internationally. Another advantage is that all patients were treated and followed up at the same institution and according to the same protocol. Furthermore, our study included patients from a genetically homogeneous Slovenian population, allowing us to exclude the influence of genetic heterogeneity on our results. On the other hand, this may limit the translation of our findings to other populations. Consequently, we were unable to directly compare our results with those from other studies, which investigated other cancers and were mostly conducted on Asian populations.

Despite being a rare cancer, MM remains a major public health concern due to high mortality and rising incidence among older, asbestos-exposed individuals, unlike younger individuals, in whom incidence is declining.^1,6,20 Therefore, it is crucial that further studies clarify the mechanisms influencing the course and treatment of MM. To our knowledge, our study is the first worldwide to investigate the role of HIF1A polymorphisms in patients with MM. Additionally, it is one of the few studies examining HIF1A polymorphisms in cancer within a European population, as most previous studies has been conducted in Asian populations.

Although we found a statistically significant poorer response to chemotherapy in heterozygous patients with the rs11549465 CT genotype after adjustment for weight loss and CRP, we believe that the results of this study do not provide new findings with immediate clinical applicability. Consequently, testing MM patients for these three HIF1A polymorphisms has for now, in our opinion, no clinical utility. Nevertheless, our findings may contribute to understanding the role of HIF1A polymorphisms in MM and cancer more broadly, but further studies will be necessary to clarify their role. Moreover, since our study included only Slovenian patients, it remains necessary to investigate whether any associations exist between HIF1A polymorphisms and chemotherapy response or survival in patients with MM in other populations, particularly non-Asian groups for which data are currently lacking. Further research on HIF1A polymorphisms is also needed to more precisely explain their cellular mechanisms and roles in various pathological processes in the body.

This study is the first to investigate the role of HIF1A polymorphisms in MM in the context of platinum-based chemotherapy. Among the variants examined, only the rs11549465 CT genotype was associated with a significantly poorer response to chemotherapy after adjusting for clinical factors. No associations were found between any of the investigated SNPs and OS or PFS. Although our findings do not currently support clinical implementation of HIF1A genotyping in MM management, they contribute to the growing body of evidence on the role of hypoxia-related genetic variability in chemotherapy resistance. Further research in larger and more diverse populations is warranted to validate these findings and explore their broader implications in cancer biology and treatment personalization.