Advertisement

Severe hypoxia is a typical characteristic of human hepatocellular carcinoma: Scientific fact or fallacy?

  • Thorsten Cramer
    Correspondence
    Corresponding authors. Addresses: Department of Surgery, Pauwelsstrasse 30, 52074 Aachen, Germany; Tel.: +49-241-80-36353, fax: +49-241-80-82068;
    Affiliations
    Department of General, Visceral and Transplantation Surgery, RWTH University Hospital, 52074 Aachen, Germany

    European Surgery Center Aachen Maastricht, Aachen, Germany

    European Surgery Center Aachen Maastricht, Maastricht, The Netherlands
    Search for articles by this author
  • Peter Vaupel
    Correspondence
    Department of Radiation Oncology, University Medical Center, University of Freiburg, Robert-Koch-Strasse 3, 79106 Freiburg/Brsg., Germany; Tel.: +49-171-1240073, fax: +49-6131-496894.
    Affiliations
    Department of Radiation Oncology, University Medical Center, University of Freiburg, Freiburg, Germany

    German Cancer Center Consortium (DKTK), Partner Site Freiburg, German Cancer Research Center (DKFZ), Heidelberg, Germany
    Search for articles by this author
Open AccessPublished:January 02, 2022DOI:https://doi.org/10.1016/j.jhep.2021.12.028

      Summary

      Hepatocellular carcinoma (HCC) is characterised by a robust resistance to therapy, resulting in the very poor prognosis usually seen in patients with unresectable HCC. A thorough understanding of the molecular and cellular pathogenesis of HCC is of paramount importance for the identification of more effective treatment options. As hypoxia in tumours is associated with the malignant phenotype, molecules involved in the hypoxic response are being investigated as potential targets for cancer therapy. One key hallmark of human HCC is the hypervascularisation and arterialisation of the tumour’s blood supply. Hypoxia being a strong inducer of neo-angiogenesis, it was hypothesised over 20 years ago that reduced oxygen levels in human HCC are a crucial feature of this deadly disease. However, while there is a considerable body of literature espousing the presumed functional relevance of hypoxia in HCC, direct measurements of oxygen partial pressures or O2 concentrations in human HCCs have yet to be performed. This narrative review seeks to demonstrate how overinterpretation of in vitro experiments and incorrect citations have resulted in HCCs being perceived as severely hypoxic tumours.

      Keywords

      Introduction

      Hepatocellular carcinoma (HCC) represents the fifth most common malignant tumour and the third most common cause of cancer-related deaths in the world. The incidence of HCC is constantly rising, mostly due to the robust increase of HCCs as a consequence of metabolic liver disease.
      • Llovet J.M.
      • Kelley R.K.
      • Villanueva A.
      • Singal A.G.
      • Pikarsky E.
      • Roayaie S.
      • et al.
      Hepatocellular carcinoma.
      At least half of all HCC cases are diagnosed at advanced tumour stages, precluding potentially curative treatment options such as surgical resection or orthotopic liver transplantation.
      • Llovet J.M.
      • Kelley R.K.
      • Villanueva A.
      • Singal A.G.
      • Pikarsky E.
      • Roayaie S.
      • et al.
      Hepatocellular carcinoma.
      The non-surgical means of treatment (e.g., chemo-, radio- and immuno-therapy or combinations thereof) are usually restricted by major pathophysiological parameters determining the tumour microenvironment (TME, e.g., limited/chaotic blood flow, critically reduced oxygen supply, severe tissue acidosis, etc.). A detailed understanding of the molecular and cellular pathogenesis of HCC is critical to improving the treatment of this deadly disease. Against this background, it is important to acknowledge the impressive accumulation of knowledge that we have witnessed in recent years. The advent of omics technologies and widely available tissue cohorts as well as data repositories have resulted in the identification of molecular HCC subtypes with differing sensitivities to therapy and prognoses.
      • Faivre S.
      • Rimassa L.
      • Finn R.S.
      Molecular therapies for HCC: looking outside the box.
      In addition, a better molecular and cellular understanding of the phenomenon of immune evasion has led to the approval of a new combination therapy (atezolizumab plus bevacizumab) for unresectable HCCs.
      • Finn R.S.
      • Qin S.
      • Ikeda M.
      • Galle P.R.
      • Ducreux M.
      • Kim T.Y.
      • et al.
      Atezolizumab plus bevacizumab in unresectable hepatocellular carcinoma.
      These success stories demonstrate the pivotal importance of well-designed, properly controlled and critically executed pre-clinical research, particularly with respect to the translation of crucial findings into clinical practice. In this review, we will outline an unfortunate, negative example, namely the (as yet unproven) biological significance of severe hypoxia in human HCCs. As briefly stated above, critically reduced oxygen (O2) supply in the TME results in aggressive disease, tumour immune evasion and reduced therapeutic efficacy. Hence, characterising hypoxia and its molecular consequences in the TME is of great significance. As HCCs are typically hypervascularised tumours and hypoxia is a well-established stimulus of neo-angiogenesis, it was hypothesised early on that a lack of oxygen in the TME of HCCs is causally involved in the pathogenesis of HCC.
      • Bae M.K.
      • Kwon Y.W.
      • Kim M.S.
      • Bae S.K.
      • Bae M.H.
      • Lee Y.M.
      • et al.
      Identification of genes differentially expressed by hypoxia in hepatocellular carcinoma cells.
      Interestingly, the physical presence of severe (= extreme) hypoxia with oxygen partial pressures (pO2 values) below 1-2 mmHg in human HCC has never been convincingly demonstrated, i.e. pO2 distributions in human HCC have never been directly and accurately assessed with appropriate tools such as minimally invasive polarographic microsensors, optical microprobes
      • Vaupel P.
      • Flood A.B.
      • Swartz H.M.
      Oxygenation status of malignant tumors vs. Normal tissues: critical evaluation and updated data source based on direct measurements with pO2 microsensors.
      or electron paramagnetic resonance oximetry with implantable oxygen probes
      • Höckel M.
      • Vaupel P.
      Tumor hypoxia: definitions and current clinical, biologic, and molecular aspects.
      . Moreover, the imaging techniques that are currently used (some of them being vascular-/blood-based) do not permit the direct quantification of tissue pO2 values and thus may not identify tissue hypoxia per se (discussed in detail below). Another method that can directly estimate the tumour oxygenation status is 19F-oximetry, which is currently in clinical evaluation.
      • D'Alonzo R.A.
      • Gill S.
      • Rowshanfarzad P.
      • Keam S.
      • MacKinnon K.M.
      • Cook A.M.
      • et al.
      In vivo noninvasive preclinical tumor hypoxia imaging methods: a review.
      The pO2 polarography method (the “established gold standard”
      • Vaupel P.
      • Schlenger K.
      • Knoop C.
      • Höckel M.
      Oxygenation of human tumors: evaluation of tissue oxygen distribution in breast cancers by computerized O2 tension measurements.
      ) has been the preferred technique for monitoring tissue pO2 values. In this method, intratumoral (intratissue) polarographic measurements adhere to the systematic random sampling principle based on linear microelectrode tracks, resulting from computerised, stepwise movements into the tissue of interest.
      • Vaupel P.
      • Höckel M.
      • Mayer A.
      Detection and characterization of tumor hypoxia using pO2 histography.
      After pO2 determination, core biopsies are taken from the tumour sub-volumes where the pO2 measurements were made, and subsequently processed for histology (exclusion of necrosis) and immunohistochemistry. Nevertheless, as of November 2021, the search terms "hepatoma and hypoxia" and "hepatocellular carcinoma and hypoxia" yielded 1,698 and 1,480 results, respectively, on PubMed. How is it, then, that the notion of “severe hypoxia in human HCC” is so widely espoused in the scientific literature? Intrigued by this question, we performed a thorough investigation of the available literature and found a combination of data misinterpretation, incorrect citation practice, inappropriate translation, and continuous perseveration of this idea to be largely responsible.

      Blood supply of hepatocellular carcinomas and secondary liver tumours

      The normal liver is well vascularised, receiving approx. 26% of cardiac output (∼1.1 ml/g/min) and consuming approx. 20% of the total O2 used by the body at rest (∼0.06 ml O2/g/min). About 75% of the hepatic blood supply is received by the portal vein, the rest by the hepatic artery. In general, hepatocarcinogenesis is associated with marked hypervascularity and increased arterialisation (arterial inflow), stimulated by various molecular pathways, including hypoxia-inducible factor (HIF)-dependent pro-angiogenic factors.
      • Morse M.A.
      • Sun W.
      • Kim R.
      • He A.R.
      • Abada P.B.
      • Mynderse M.
      • et al.
      The role of angiogenesis in hepatocellular carcinoma.
      As a consequence of the highly increased arterialisation, HCCs show a significantly higher arterial blood flow and arterial fraction (50-60%) compared to normal liver parenchyma (∼25%). Overall, HCCs typically present with a dominating (or nearly exclusive) arterial blood supply.
      • Morse M.A.
      • Sun W.
      • Kim R.
      • He A.R.
      • Abada P.B.
      • Mynderse M.
      • et al.
      The role of angiogenesis in hepatocellular carcinoma.
      In secondary liver tumours (e.g., liver metastases from colorectal cancer), neo-angiogenesis is often dysfunctional and inadequate, resulting in chaotic and heterogeneous blood flow at early growth stages.
      • Burke D.
      • Davies M.M.
      • Zweit J.
      • Flower M.A.
      • Ott R.J.
      • Dworkin M.J.
      • et al.
      Continuous angiotensin II infusion increases tumour: normal blood flow ratio in colo-rectal liver metastases.
      ,
      • Forner A.
      • Ayuso C.
      • Bruix J.
      Hepatocellular Carcinoma. Schiff's Diseases of the Liver (Sorrell, MF; Maddrey, WC; Schiff, ER).
      In normal liver parenchyma, the median pO2 ranges between 30 and 40 mmHg (for a recent review see
      • Vaupel P.
      • Flood A.B.
      • Swartz H.M.
      Oxygenation status of malignant tumors vs. Normal tissues: critical evaluation and updated data source based on direct measurements with pO2 microsensors.
      ). In the mid-1990s, direct measurements in liver metastases of rectal cancer yielded a median pO2 of 6 mmHg,
      • Kallinowski F.
      • Buhr H.J.
      Tissue oxygenation of primary, metastatic and xenografted rectal cancers.
      the fraction of hypoxic pO2 values ≤10 mmHg being 75%.
      • Kallinowski F.
      • Buhr H.J.
      Tissue oxygenation of primary, metastatic and xenografted rectal cancers.
      ,
      • Kallinowski F.
      • Buhr H.J.
      Can the oxygenation status of rectal carcinomas be improved by hypoxia?.
      Hence, the assumption that hypoxia plays a functional role in the induction of neo-angiogenesis in liver metastases is based on direct measurements of oxygen in these tissues. In contrast, direct measurements of pO2 in human HCC have not yet been published, thus the functional relevance of hypoxia to hypervascularity and arterialisation of HCC remains unsubstantiated.

      Hypoxia, a hallmark of cancer and an independent, adverse prognostic factor in human malignancies

      In general, tissue hypoxia characterises a (critically) deficient oxygenation status (pO2 ≤10 mmHg) and is distributed heterogeneously within and across tumours.
      • Vaupel P.
      • Höckel M.
      • Mayer A.
      Detection and characterization of tumor hypoxia using pO2 histography.
      Severe hypoxia is not yet well defined, i.e., there is no generally accepted standard or sharp threshold for quantifying severe hypoxia in the clinical setting. Based on several experimental and clinical observations, pO2 values below 1–2 mmHg are deemed indicative of the existence of severe hypoxia in tumour sub-volumes.
      • Höckel M.
      • Vaupel P.
      Tumor hypoxia: definitions and current clinical, biologic, and molecular aspects.
      Spatial and temporal heterogeneities of tissue hypoxia on micro- and macro-vascular scales, as well as pronounced differences in the severity (extent), size and distribution of the respective sub-volumes, are principal hallmarks of most cancer types,
      • Semenza G.L.
      The hypoxic tumor microenvironment: a driving force for breast cancer progression.
      directly affecting or even causing many other hallmarks.
      • Vaupel P.
      • Flood A.B.
      • Swartz H.M.
      Oxygenation status of malignant tumors vs. Normal tissues: critical evaluation and updated data source based on direct measurements with pO2 microsensors.
      Since the early 1990s, hypoxia has been recognised as a detrimental condition, which in turn critically influences a variety of biological responses that increase malignant potential through changes in the genome, transcriptome and proteome.
      • Harris A.L.
      Hypoxia--a key regulatory factor in tumour growth.
      In addition, hypoxia can act as a barrier to the effectiveness of oxygen-dependent treatment modalities (e.g., standard radiotherapy, some chemotherapeutic agents, immunotherapy and photodynamic therapy). Taken together, tumour hypoxia is associated with poor patient prognosis. First described by Höckel and Vaupel,
      • Vaupel P.
      • Schlenger K.
      • Knoop C.
      • Höckel M.
      Oxygenation of human tumors: evaluation of tissue oxygen distribution in breast cancers by computerized O2 tension measurements.
      ,
      • Höckel M.
      • Schlenger K.
      • Knoop C.
      • Vaupel P.
      Oxygenation of carcinomas of the uterine cervix: evaluation by computerized O2 tension measurements.
      this association has since been endorsed by Gregg L. Semenza (e.g.
      • Semenza G.L.
      The hypoxic tumor microenvironment: a driving force for breast cancer progression.
      ). Systematic direct assessment of the oxygenation status and detection of hypoxia in primary cancers, metastases and recurrent malignancies were initiated in the late 1980s by the Höckel/Vaupel group (for a review see
      • Vaupel P.
      • Flood A.B.
      • Swartz H.M.
      Oxygenation status of malignant tumors vs. Normal tissues: critical evaluation and updated data source based on direct measurements with pO2 microsensors.
      ). These clinical studies were expanded by correlation analyses of directly measured pO2 values and expression of endogenous hypoxia markers in the Vaupel/Mayer laboratory.
      • Mayer A.
      • Höckel M.
      • Wree A.
      • Vaupel P.
      Microregional expression of glucose transporter-1 and oxygenation status: lack of correlation in locally advanced cervical cancers.
      • Mayer A.
      • Wree A.
      • Höckel M.
      • Leo C.
      • Pilch H.
      • Vaupel P.
      Lack of correlation between expression of HIF-1alpha protein and oxygenation status in identical tissue areas of squamous cell carcinomas of the uterine cervix.
      • Mayer A.
      • Höckel M.
      • Vaupel P.
      Carbonic anhydrase IX expression and tumor oxygenation status do not correlate at the microregional level in locally advanced cancers of the uterine cervix.
      The prerequisites for these latter coexistence analyses were needle core biopsies immediately after the pO2 measurements, as mentioned earlier in this article. These core biopsies, including the tumour sub-volume around the microsensor track allowed (a) exclusions of pO2 measurements in necrotic regions, and (b) possible correlations between endogenous biomarkers and the directly assessed oxygenation status.
      • Mayer A.
      • Höckel M.
      • Wree A.
      • Vaupel P.
      Microregional expression of glucose transporter-1 and oxygenation status: lack of correlation in locally advanced cervical cancers.
      • Mayer A.
      • Wree A.
      • Höckel M.
      • Leo C.
      • Pilch H.
      • Vaupel P.
      Lack of correlation between expression of HIF-1alpha protein and oxygenation status in identical tissue areas of squamous cell carcinomas of the uterine cervix.
      • Mayer A.
      • Höckel M.
      • Vaupel P.
      Carbonic anhydrase IX expression and tumor oxygenation status do not correlate at the microregional level in locally advanced cancers of the uterine cervix.
      Disappointingly, however, the expression of HIF-1α and its downstream proteins (e.g., GLUT-1, CAIX) did not correlate with directly assessed oxygenation status. In addition, a hypoxic response in uterine leiomyomas, despite severe tissue hypoxia, was never observed.
      • Mayer A.
      • Höckel M.
      • Wree A.
      • Leo C.
      • Horn L.C.
      • Vaupel P.
      Lack of hypoxic response in uterine leiomyomas despite severe tissue hypoxia.
      These data are hardly surprising, given that the expression of HIF-1α and its downstream proteins is also triggered by non-hypoxic signalling
      • LaGory E.L.
      • Giaccia A.J.
      The ever-expanding role of HIF in tumour and stromal biology.
      (e.g., in HCCs strongly relying on the Warburg effect and distinctly accelerated metabolism). It is important to note that, unless the detection of the immunohistochemical biomarker and the direct pO2 readings are made on the basis of identical tumour sub-volumes and/or tumour regions, the correlation analyses based on these two data sets will be irrelevant. To the best of our knowledge, this requirement has never been met for HCCs.

      Critical evaluation of data on the existence of “severe hypoxia in HCC”

      In light of this, it is difficult to imagine that the presence and functional significance of severe hypoxia during the early pathogenesis of HCC can play a decisive role. A logical first step in an experimental investigation on the role of hypoxia in HCC would be to directly measure the partial pressure (or concentration) of oxygen and compare it to normal liver and HCC precursors such as cirrhosis and steatohepatitis. The two studies that involve direct measurements of pO2 in rodent HCC models provide conflicting results. Hypoxia was detected in a rat hepatoma model (heterotopically transplanted into skeletal muscle),
      • Cater D.B.
      • Schoeniger E.L.
      • Watkinson D.A.
      Effect of breathing high pressure oxygen upon tissue oxygen tension in rat and mouse tumours.
      while HCC nodules of the well-established and widely used diethylnitrosamine model failed to demonstrate reduced pO2 compared to normal murine liver.
      • Tanaka H.
      • Yamamoto M.
      • Hashimoto N.
      • Miyakoshi M.
      • Tamakawa S.
      • Yoshie M.
      • et al.
      Hypoxia-independent overexpression of hypoxia-inducible factor 1alpha as an early change in mouse hepatocarcinogenesis.
      These pioneering researchers used microelectrodes (i.e., the “polarographic oxygen cathode technique”), which enabled direct measurements of oxygen tensions inside the tumours. Decades of subsequent research used the established rodent and human HCC cell lines primarily as tools to analyse the cellular hypoxic response (see note above). These experiments, among others, led to a detailed characterisation of the cellular response to hypoxia
      • Arany Z.
      • Huang L.E.
      • Eckner R.
      • Bhattacharya S.
      • Jiang C.
      • Goldberg M.A.
      • et al.
      An essential role for p300/CBP in the cellular response to hypoxia.
      ,
      • Goldberg M.A.
      • Glass G.A.
      • Cunningham J.M.
      • Bunn H.F.
      The regulated expression of erythropoietin by two human hepatoma cell lines.
      as well as the identification of HIF-1 (G. Wang and G. L. Semenza
      • Semenza G.L.
      • Wang G.L.
      A nuclear factor induced by hypoxia via de novo protein synthesis binds to the human erythropoietin gene enhancer at a site required for transcriptional activation.
      ,
      • Wang G.L.
      • Semenza G.L.
      Characterization of hypoxia-inducible factor 1 and regulation of DNA binding activity by hypoxia.
      ). While these discoveries were, and continue to be, of paramount importance to biology, the functional role of hypoxia in the pathogenesis of (rodent or human) HCC in vivo remains unexplored. Interestingly, the aforementioned publication from 1963, demonstrating reduced partial pressure of oxygen in a rat HCC model,
      • Cater D.B.
      • Schoeniger E.L.
      • Watkinson D.A.
      Effect of breathing high pressure oxygen upon tissue oxygen tension in rat and mouse tumours.
      remains the only direct proof of hypoxia in HCC and has not, to date, been confirmed by studies on human HCC.
      • Xiong X.X.
      • Qiu X.Y.
      • Hu D.X.
      • Chen X.Q.
      Advances in hypoxia-mediated mechanisms in hepatocellular carcinoma.
      In this context, it may be worthwhile to briefly outline the circumstances that led to the widespread designation of human HCC as a "typically hypoxic tumour" (“severely hypoxic tumour”) (see also timeline in Fig. 1). As mentioned above, HCC is a hypervascularised tumour and hypoxia (through the induction of gene expression of various pro-angiogenic growth factors) is a strong inducer of new blood vessel formation (neo-angiogenesis). This fact led the group headed by B.C. Park and K.W. Kim to hypothesise that hypoxia induces angiogenesis in human HCC, resulting in the hypervascularised HCC phenotype and tumour progression.
      • Kim K.R.
      • Moon H.E.
      • Kim K.W.
      Hypoxia-induced angiogenesis in human hepatocellular carcinoma.
      This postulation was based on the original finding by Park et al. of elevated gene and protein expression of the pro-angiogenic factor insulin-like growth factor II (IGF-II) in human HCC.
      • Park B.C.
      • Huh M.H.
      • Seo J.H.
      Differential expression of transforming growth factor alpha and insulin-like growth factor II in chronic active hepatitis B, cirrhosis and hepatocellular carcinoma.
      Subsequently, the research team led by K.W. Kim reported that hypoxia in vitro is able to activate gene expression of IGF-II and other pro-angiogenic factors in established human HCC cell lines,
      • Bae M.K.
      • Kwon Y.W.
      • Kim M.S.
      • Bae S.K.
      • Bae M.H.
      • Lee Y.M.
      • et al.
      Identification of genes differentially expressed by hypoxia in hepatocellular carcinoma cells.
      ,
      • Kim K.W.
      • Bae S.K.
      • Lee O.H.
      • Bae M.H.
      • Lee M.J.
      • Park B.C.
      Insulin-like growth factor II induced by hypoxia may contribute to angiogenesis of human hepatocellular carcinoma.
      concluding that “HCC … may have substantial hypoxic fractions”
      • Bae M.K.
      • Kwon Y.W.
      • Kim M.S.
      • Bae S.K.
      • Bae M.H.
      • Lee Y.M.
      • et al.
      Identification of genes differentially expressed by hypoxia in hepatocellular carcinoma cells.
      and that “hypoxia may play a key role in … hepatocarcinogenesis”.
      • Kim K.W.
      • Bae S.K.
      • Lee O.H.
      • Bae M.H.
      • Lee M.J.
      • Park B.C.
      Insulin-like growth factor II induced by hypoxia may contribute to angiogenesis of human hepatocellular carcinoma.
      This cautious (and in our opinion correct) assessment was replaced by a rather unequivocal assertion in a subsequent article by the same group: “Human solid tumours, including HCC … have substantial hypoxic fractions.”.
      • Lee S.W.
      • Lee Y.M.
      • Bae S.K.
      • Murakami S.
      • Yun Y.
      • Kim K.W.
      Human hepatitis B virus X protein is a possible mediator of hypoxia-induced angiogenesis in hepatocarcinogenesis.
      Strikingly, the referenced article by Moulder and Rockwell
      • Moulder J.E.
      • Rockwell S.
      Hypoxic fractions of solid tumors: experimental techniques, methods of analysis, and a survey of existing data.
      addresses the role of hypoxia in rodent tumour models alone with no mention whatsoever of human HCC, demonstrating the citation’s inaccuracy in this context.
      Figure thumbnail gr1
      Fig. 1Graphical depiction of the central findings and publications with respect to the role of severe hypoxia in human HCC.
      PubMed results for the search term “hepatocellular carcinoma and hypoxia” from 1990 to 2020 are shown in green. Articles with correct data presentation/interpretation are shown in light green, those with incorrect data citation/ interpretation are in red boxes. HCC, hepatocellular carcinoma; pO2, partial pressure of oxygen.
      Further investigation revealed multiple incorrect references to a review published by one of us (PV,
      • Vaupel P.
      • Höckel M.
      • Mayer A.
      Detection and characterization of tumor hypoxia using pO2 histography.
      ) pertaining to the detection methods of tumour hypoxia. In this article, an overview of the oxygenation status of miscellaneous human tumours is presented in table format. Despite its use of precise and unambiguous labelling ("Liver tumours (metastatic)”), we found the article to have been incorrectly cited by several research groups
      • Chiu D.K.
      • Tse A.P.
      • Xu I.M.
      • Di Cui J.
      • Lai R.K.
      • Li L.L.
      • et al.
      Hypoxia inducible factor HIF-1 promotes myeloid-derived suppressor cells accumulation through ENTPD2/CD39L1 in hepatocellular carcinoma.
      • Hu W.
      • Zheng S.
      • Guo H.
      • Dai B.
      • Ni J.
      • Shi Y.
      • et al.
      PLAGL2-EGFR-HIF-1/2α signaling loop promotes HCC progression and Erlotinib insensitivity.
      • Kuchuk O.
      • Tuccitto A.
      • Citterio D.
      • Huber V.
      • Camisaschi C.
      • Milione M.
      • et al.
      pH regulators to target the tumor immune microenvironment in human hepatocellular carcinoma.
      • Lin Z.
      • Niu Y.
      • Wan A.
      • Chen D.
      • Liang H.
      • Chen X.
      • et al.
      RNA m(6) A methylation regulates sorafenib resistance in liver cancer through FOXO3-mediated autophagy.
      • Wong C.C.-L.
      • Kai A.K.-L.
      • Ng I.O.-L.
      The impact of hypoxia in hepatocellular carcinoma metastasis.
      • Yuen V.W.
      • Wong C.C.
      Hypoxia-inducible factors and innate immunity in liver cancer.
      as proof of (severe) hypoxia in HCC. Likewise, a review on hypoxia in malignant tumours by S. R. McKeown is commonly used to support the notion of severe hypoxia as a well-established characteristic of HCC.
      • McKeown S.R.
      Defining normoxia, physoxia and hypoxia in tumours-implications for treatment response.
      Interestingly, however, while this review mentions the evidence of hypoxia (6 mmHg median tumour pO2 compared to 30 mmHg median pO2 in normal liver tissue) in the tumour type "liver" (Table 2 in
      • McKeown S.R.
      Defining normoxia, physoxia and hypoxia in tumours-implications for treatment response.
      ), the original literature cited in support of this statement did not determine the pO2 in human HCC, but in liver metastases of rectal cancer.
      • Kallinowski F.
      • Buhr H.J.
      Can the oxygenation status of rectal carcinomas be improved by hypoxia?.
      ,
      • Semenza G.L.
      The hypoxic tumor microenvironment: a driving force for breast cancer progression.
      Thus, the particular use of the terms "liver" and "tumour type" in McKeown’s review ("secondary liver tumour" would have been more accurate) led the authors of several articles to (incorrectly) conclude that hypoxia is a typical feature of HCC.
      • Bhattacharya B.
      • Huang D.Q.
      • Low S.H.H.
      • Tan G.H.
      • Han M.J.
      • Singh S.
      • et al.
      Effect of cell microenvironment on the drug sensitivity of hepatocellular cancer cells.
      • Cai J.
      • Hu M.
      • Chen Z.
      • Ling Z.
      The roles and mechanisms of hypoxia in liver fibrosis.
      • Chen C.
      • Lou T.
      Hypoxia inducible factors in hepatocellular carcinoma.
      • Sheng J.
      • Qin H.
      • Zhang K.
      • Li B.
      • Zhang X.
      Targeting autophagy in chemotherapy-resistant of hepatocellular carcinoma.
      • Su Q.
      • Fan M.
      • Wang J.
      • Ullah A.
      • Ghauri M.A.
      • Dai B.
      • et al.
      Sanguinarine inhibits epithelial-mesenchymal transition via targeting HIF-1α/TGF-β feed-forward loop in hepatocellular carcinoma.
      The review by McKeown had several other flaws: (a) the lack of correct units for pO2 (% instead of mmHg, Torr or kPa), (b) inappropriate use of Henry’s law and Dalton’s law (both are only valid for gas mixtures or gases dissolved in homogeneous solutions), and (c) the assumption that Bunsen’s solubility coefficient α and Henry’s law constant H are valid for heterogeneous tissues (Note: α greatly depends on the water and lipid contents of a tissue, the volume of the extravascular space, etc.; for further details see
      • Keeley T.P.
      • Mann G.E.
      Defining physiological normoxia for improved translation of cell physiology to animal models and humans.
      ). We found a number of additional fallacies in the literature on hypoxia and HCC, e.g., the assertion “hypoxia is typical for HCC” being based on references to reviews that do not mention HCC at all, and other claims without citing relevant sources. However, as discrediting individual researchers is not the purpose of this review, we decided not to list all of them here.
      It is conceivable that non-invasive imaging methods (mentioned in the Introduction) were able to demonstrate severe hypoxia in human HCC. Indeed, blood-oxygenation-level-dependent (BOLD)-MRI and tissue-oxygenation-level-dependent (TOLD)-MRI were used in the late 2010s to characterise the oxygenation status of human HCCs. These investigations are indicative of (a) pronounced oxygenation heterogeneities within and between HCCs,
      • Hectors S.J.
      • Wagner M.
      • Bane O.
      • Besa C.
      • Lewis S.
      • Remark R.
      • et al.
      Quantification of hepatocellular carcinoma heterogeneity with multiparametric magnetic resonance imaging.
      (b) a poorer oxygenation status in HCCs compared to adjacent normal liver tissues with evidence of an arterialisation of the blood supply to the cancers,
      • Patterson A.J.
      • Priest A.N.
      • Bowden D.J.
      • Wallace T.E.
      • Patterson I.
      • Graves M.J.
      • et al.
      Quantitative BOLD imaging at 3T: temporal changes in hepatocellular carcinoma and fibrosis following oxygen challenge.
      (c) comparable R2∗ values (i.e., comparable oxygenation) when investigating HCCs compared to skeletal muscle during air breathing,
      • Bane O.
      • Besa C.
      • Wagner M.
      • Oesingmann N.
      • Zhu H.
      • Fiel M.I.
      • et al.
      Feasibility and reproducibility of BOLD and TOLD measurements in the liver with oxygen and carbogen gas challenge in healthy volunteers and patients with hepatocellular carcinoma.
      (d) declining T2∗ values when assessing liver haemangiomas, HCCs, metastatic liver tumours and cholangiocarcinomas, i.e., deteriorating oxygenation status in this order,
      • Park H.J.
      • Kim Y.K.
      • Min J.H.
      • Lee W.J.
      • Choi D.
      • Rhim H.
      Feasibility of blood oxygenation level-dependent MRI at 3T in the characterization of hepatic tumors.
      and (e) the oxygenation status of HCCs being superior to that of liver metastases and cholangiocarcinomas.
      • Park H.J.
      • Kim Y.K.
      • Min J.H.
      • Lee W.J.
      • Choi D.
      • Rhim H.
      Feasibility of blood oxygenation level-dependent MRI at 3T in the characterization of hepatic tumors.
      None of these articles (a-e) reported “severe hypoxia” in HCCs.

      The applicability of gene expression signatures to detect hypoxia in tissues

      Omics technologies and bioinformatic analysis tools have substantially broadened our understanding of HCC pathophysiology, leading to the establishment of molecular HCC subtypes.
      • Llovet J.M.
      • Kelley R.K.
      • Villanueva A.
      • Singal A.G.
      • Pikarsky E.
      • Roayaie S.
      • et al.
      Hepatocellular carcinoma.
      The hypoxia-induced gene expression signatures that have been identified thus far demonstrate prognostic relevance in various cancer types.
      • Buffa F.M.
      • Harris A.L.
      • West C.M.
      • Miller C.J.
      Large meta-analysis of multiple cancers reveals a common, compact and highly prognostic hypoxia metagene.
      At times, these signatures are used to deduce the presence of hypoxia in tissues, e.g., for human HCC.
      • Ma L.
      • Hernandez M.O.
      • Zhao Y.
      • Mehta M.
      • Tran B.
      • Kelly M.
      • et al.
      Tumor cell biodiversity drives microenvironmental reprogramming in liver cancer.
      Let us briefly outline why this is not a valid experimental approach. The cellular response to hypoxia is generally mediated by the transcription factors HIF-1 and HIF-2 (henceforth termed HIF).
      • Schito L.
      • Semenza G.L.
      Hypoxia-inducible factors: master regulators of cancer progression.
      Hundreds of HIF target genes have been identified and are used to characterise the transcriptional response to hypoxia. The term “hypoxia” in bioinformatics-based gene set and pathway analyses is largely contingent on the activation of HIF target genes. It is of pivotal importance to understand that the stabilisation and activation of the HIF proteins are not restricted to hypoxia, but, as briefly mentioned above, can be achieved by various hypoxia-independent means (Fig. 2).
      • Iommarini L.
      • Porcelli A.M.
      • Gasparre G.
      • Kurelac I.
      Non-canonical mechanisms regulating hypoxia-inducible factor 1 alpha in cancer.
      To appreciate this, the term “non-canonical regulation of HIF signaling” was introduced by Amato Giaccia and colleagues in 2016.
      • LaGory E.L.
      • Giaccia A.J.
      The ever-expanding role of HIF in tumour and stromal biology.
      In light of this, it is inherently impossible to use HIF target gene activation as a proxy for tissue hypoxia as this approach completely neglects the non-canonical means of HIF stabilisation.
      Figure thumbnail gr2
      Fig. 2Experimental workflow and bioinformatic algorithm based on a “HIF gene expression signature”.
      Canonical (mediated by hypoxia) and non-canonical (mediated by various hypoxia-independent means as depicted) HIF stabilisation. Bioinformatic tools (green arrows) are able to identify HIF activity based on expression of HIF target genes. Of note, the underlying route to HIF activity (hypoxic vs. hypoxia-independent) cannot be deduced from this algorithm. HIF, hypoxia-inducible factor.

      Conclusions

      In this narrative review, we have sought to clarify that there is scant evidence to support the notion of severe hypoxia in human HCCs as the relevant results are inconclusive.
      • Morse M.A.
      • Sun W.
      • Kim R.
      • He A.R.
      • Abada P.B.
      • Mynderse M.
      • et al.
      The role of angiogenesis in hepatocellular carcinoma.
      ,
      • Xiong X.X.
      • Qiu X.Y.
      • Hu D.X.
      • Chen X.Q.
      Advances in hypoxia-mediated mechanisms in hepatocellular carcinoma.
      The widespread remark “hypoxia is a typical feature of HCC” is not supported by reliable, published data and should not be used. In order for severe hypoxia to be considered a hallmark of this tumour type, there must be incontrovertible proof of its presence in most HCCs. No argument to the contrary would be scientifically sound. Adequately referencing the published literature can be a daunting, time-intensive task. The problem discussed here is an example of how the ever-increasing pressure to produce and publish research articles can not only affect the diligence required for meticulous referencing but also, more crucially, obscure the scientific facts, resulting in publications of questionable merit and little value to the scientific community. Certain characteristics of human HCC, e.g., hypervascularity, areas of necrotic tumour tissue and robust primary resistance to therapy, do indicate the presence of severe hypoxia in this tumour type. We therefore maintain that it is only an indisputable demonstration of severely reduced pO2 in human HCC that can ascertain the functional relevance of hypoxia for this tumour type in humans.

      Abbreviations

      HCC, hepatocellular carcinoma; HIF, hypoxia-inducible factor; IGF-II, insulin-like growth factor II; pO2, partial pressure of oxygen; TME, tumour microenvironment.

      Financial support

      The authors received no financial support to produce this manuscript.

      Authors' contributions

      TC and PV performed literature research and wrote the manuscript.

      Conflict of interest

      The authors have no conflicts of interest to declare.
      Please refer to the accompanying ICMJE disclosure forms for further details.

      Acknowledgments

      We thank Eugene Datta for editing the text of the manuscript and Antje Egners for graphical assistance. Thorsten Cramer is supported by a grant from the Deutsche Forschungsgemeinschaft ( CR 133/5-1 ).

      Supplementary data

      The following is the supplementary data to this article:

      References

        • Llovet J.M.
        • Kelley R.K.
        • Villanueva A.
        • Singal A.G.
        • Pikarsky E.
        • Roayaie S.
        • et al.
        Hepatocellular carcinoma.
        Nat Rev Dis Primers. 2021; 7: 6
        • Faivre S.
        • Rimassa L.
        • Finn R.S.
        Molecular therapies for HCC: looking outside the box.
        J Hepatol. 2020; 72: 342-352
        • Finn R.S.
        • Qin S.
        • Ikeda M.
        • Galle P.R.
        • Ducreux M.
        • Kim T.Y.
        • et al.
        Atezolizumab plus bevacizumab in unresectable hepatocellular carcinoma.
        N Engl J Med. 2020; 382: 1894-1905
        • Bae M.K.
        • Kwon Y.W.
        • Kim M.S.
        • Bae S.K.
        • Bae M.H.
        • Lee Y.M.
        • et al.
        Identification of genes differentially expressed by hypoxia in hepatocellular carcinoma cells.
        Biochem Biophys Res Commun. 1998; 243: 158-162
        • Vaupel P.
        • Flood A.B.
        • Swartz H.M.
        Oxygenation status of malignant tumors vs. Normal tissues: critical evaluation and updated data source based on direct measurements with pO2 microsensors.
        Appl Magn Reson. 2021;
        • Höckel M.
        • Vaupel P.
        Tumor hypoxia: definitions and current clinical, biologic, and molecular aspects.
        J Natl Cancer Inst. 2001; 93: 266-276
        • D'Alonzo R.A.
        • Gill S.
        • Rowshanfarzad P.
        • Keam S.
        • MacKinnon K.M.
        • Cook A.M.
        • et al.
        In vivo noninvasive preclinical tumor hypoxia imaging methods: a review.
        Int J Radiat Biol. 2021; 97: 593-631
        • Vaupel P.
        • Schlenger K.
        • Knoop C.
        • Höckel M.
        Oxygenation of human tumors: evaluation of tissue oxygen distribution in breast cancers by computerized O2 tension measurements.
        Cancer Res. 1991; 51: 3316-3322
        • Vaupel P.
        • Höckel M.
        • Mayer A.
        Detection and characterization of tumor hypoxia using pO2 histography.
        Antioxid Redox Signal. 2007; 9: 1221-1235
        • Morse M.A.
        • Sun W.
        • Kim R.
        • He A.R.
        • Abada P.B.
        • Mynderse M.
        • et al.
        The role of angiogenesis in hepatocellular carcinoma.
        Clin Cancer Res. 2019; 25: 912-920
        • Burke D.
        • Davies M.M.
        • Zweit J.
        • Flower M.A.
        • Ott R.J.
        • Dworkin M.J.
        • et al.
        Continuous angiotensin II infusion increases tumour: normal blood flow ratio in colo-rectal liver metastases.
        Br J Cancer. 2001; 85: 1640-1645
        • Forner A.
        • Ayuso C.
        • Bruix J.
        Hepatocellular Carcinoma. Schiff's Diseases of the Liver (Sorrell, MF; Maddrey, WC; Schiff, ER).
        2011
        • Kallinowski F.
        • Buhr H.J.
        Tissue oxygenation of primary, metastatic and xenografted rectal cancers.
        in: Vaupel P. Kelleher D.K. Günderoth M. Tumor oxygenation. Gustav Fischer, Stuttgart, Germany1995: 205-209
        • Kallinowski F.
        • Buhr H.J.
        Can the oxygenation status of rectal carcinomas be improved by hypoxia?.
        in: Vaupel P. Kelleher D.K. Günderoth M. Tumor oxygenation. Gustav Fischer, Stuttgart, Germany1995: 291-296
        • Semenza G.L.
        The hypoxic tumor microenvironment: a driving force for breast cancer progression.
        Biochim Biophys Acta. 2016; 1863: 382-391
        • Harris A.L.
        Hypoxia--a key regulatory factor in tumour growth.
        Nat Rev Cancer. 2002; 2: 38-47
        • Höckel M.
        • Schlenger K.
        • Knoop C.
        • Vaupel P.
        Oxygenation of carcinomas of the uterine cervix: evaluation by computerized O2 tension measurements.
        Cancer Res. 1991; 51: 6098-6102
        • Mayer A.
        • Höckel M.
        • Wree A.
        • Vaupel P.
        Microregional expression of glucose transporter-1 and oxygenation status: lack of correlation in locally advanced cervical cancers.
        Clin Cancer Res. 2005; 11: 2768-2773
        • Mayer A.
        • Wree A.
        • Höckel M.
        • Leo C.
        • Pilch H.
        • Vaupel P.
        Lack of correlation between expression of HIF-1alpha protein and oxygenation status in identical tissue areas of squamous cell carcinomas of the uterine cervix.
        Cancer Res. 2004; 64: 5876-5881
        • Mayer A.
        • Höckel M.
        • Vaupel P.
        Carbonic anhydrase IX expression and tumor oxygenation status do not correlate at the microregional level in locally advanced cancers of the uterine cervix.
        Clin Cancer Res. 2005; 11: 7220-7225
        • Mayer A.
        • Höckel M.
        • Wree A.
        • Leo C.
        • Horn L.C.
        • Vaupel P.
        Lack of hypoxic response in uterine leiomyomas despite severe tissue hypoxia.
        Cancer Res. 2008; 68: 4719-4726
        • LaGory E.L.
        • Giaccia A.J.
        The ever-expanding role of HIF in tumour and stromal biology.
        Nat Cell Biol. 2016; 18: 356-365
        • Cater D.B.
        • Schoeniger E.L.
        • Watkinson D.A.
        Effect of breathing high pressure oxygen upon tissue oxygen tension in rat and mouse tumours.
        Acta Radiol Ther Phys Biol. 1963; 1: 233-252
        • Tanaka H.
        • Yamamoto M.
        • Hashimoto N.
        • Miyakoshi M.
        • Tamakawa S.
        • Yoshie M.
        • et al.
        Hypoxia-independent overexpression of hypoxia-inducible factor 1alpha as an early change in mouse hepatocarcinogenesis.
        Cancer Res. 2006; 66: 11263-11270
        • Arany Z.
        • Huang L.E.
        • Eckner R.
        • Bhattacharya S.
        • Jiang C.
        • Goldberg M.A.
        • et al.
        An essential role for p300/CBP in the cellular response to hypoxia.
        Proc Natl Acad Sci U S A. 1996; 93: 12969-12973
        • Goldberg M.A.
        • Glass G.A.
        • Cunningham J.M.
        • Bunn H.F.
        The regulated expression of erythropoietin by two human hepatoma cell lines.
        Proc Natl Acad Sci U S A. 1987; 84: 7972-7976
        • Semenza G.L.
        • Wang G.L.
        A nuclear factor induced by hypoxia via de novo protein synthesis binds to the human erythropoietin gene enhancer at a site required for transcriptional activation.
        Mol Cell Biol. 1992; 12: 5447-5454
        • Wang G.L.
        • Semenza G.L.
        Characterization of hypoxia-inducible factor 1 and regulation of DNA binding activity by hypoxia.
        J Biol Chem. 1993; 268: 21513-21518
        • Xiong X.X.
        • Qiu X.Y.
        • Hu D.X.
        • Chen X.Q.
        Advances in hypoxia-mediated mechanisms in hepatocellular carcinoma.
        Mol Pharmacol. 2017; 92: 246-255
        • Kim K.R.
        • Moon H.E.
        • Kim K.W.
        Hypoxia-induced angiogenesis in human hepatocellular carcinoma.
        J Mol Med (Berl). 2002; 80: 703-714
        • Park B.C.
        • Huh M.H.
        • Seo J.H.
        Differential expression of transforming growth factor alpha and insulin-like growth factor II in chronic active hepatitis B, cirrhosis and hepatocellular carcinoma.
        J Hepatol. 1995; 22: 286-294
        • Kim K.W.
        • Bae S.K.
        • Lee O.H.
        • Bae M.H.
        • Lee M.J.
        • Park B.C.
        Insulin-like growth factor II induced by hypoxia may contribute to angiogenesis of human hepatocellular carcinoma.
        Cancer Res. 1998; 58: 348-351
        • Lee S.W.
        • Lee Y.M.
        • Bae S.K.
        • Murakami S.
        • Yun Y.
        • Kim K.W.
        Human hepatitis B virus X protein is a possible mediator of hypoxia-induced angiogenesis in hepatocarcinogenesis.
        Biochem Biophys Res Commun. 2000; 268: 456-461
        • Moulder J.E.
        • Rockwell S.
        Hypoxic fractions of solid tumors: experimental techniques, methods of analysis, and a survey of existing data.
        Int J Radiat Oncol Biol Phys. 1984; 10: 695-712
        • Chiu D.K.
        • Tse A.P.
        • Xu I.M.
        • Di Cui J.
        • Lai R.K.
        • Li L.L.
        • et al.
        Hypoxia inducible factor HIF-1 promotes myeloid-derived suppressor cells accumulation through ENTPD2/CD39L1 in hepatocellular carcinoma.
        Nat Commun. 2017; 8: 517
        • Hu W.
        • Zheng S.
        • Guo H.
        • Dai B.
        • Ni J.
        • Shi Y.
        • et al.
        PLAGL2-EGFR-HIF-1/2α signaling loop promotes HCC progression and Erlotinib insensitivity.
        Hepatology. 2020;
        • Kuchuk O.
        • Tuccitto A.
        • Citterio D.
        • Huber V.
        • Camisaschi C.
        • Milione M.
        • et al.
        pH regulators to target the tumor immune microenvironment in human hepatocellular carcinoma.
        Oncoimmunology. 2018; 7e1445452
        • Lin Z.
        • Niu Y.
        • Wan A.
        • Chen D.
        • Liang H.
        • Chen X.
        • et al.
        RNA m(6) A methylation regulates sorafenib resistance in liver cancer through FOXO3-mediated autophagy.
        Embo J. 2020; 39e103181
        • Wong C.C.-L.
        • Kai A.K.-L.
        • Ng I.O.-L.
        The impact of hypoxia in hepatocellular carcinoma metastasis.
        Front Med. 2014; 8: 33-41
        • Yuen V.W.
        • Wong C.C.
        Hypoxia-inducible factors and innate immunity in liver cancer.
        J Clin Invest. 2020; 130: 5052-5062
        • McKeown S.R.
        Defining normoxia, physoxia and hypoxia in tumours-implications for treatment response.
        Br J Radiol. 2014; 87: 20130676
        • Bhattacharya B.
        • Huang D.Q.
        • Low S.H.H.
        • Tan G.H.
        • Han M.J.
        • Singh S.
        • et al.
        Effect of cell microenvironment on the drug sensitivity of hepatocellular cancer cells.
        Oncotarget. 2021; 12: 674-685
        • Cai J.
        • Hu M.
        • Chen Z.
        • Ling Z.
        The roles and mechanisms of hypoxia in liver fibrosis.
        J Transl Med. 2021; 19: 186
        • Chen C.
        • Lou T.
        Hypoxia inducible factors in hepatocellular carcinoma.
        Oncotarget. 2017; 8: 46691-46703
        • Sheng J.
        • Qin H.
        • Zhang K.
        • Li B.
        • Zhang X.
        Targeting autophagy in chemotherapy-resistant of hepatocellular carcinoma.
        Am J Cancer Res. 2018; 8: 354-365
        • Su Q.
        • Fan M.
        • Wang J.
        • Ullah A.
        • Ghauri M.A.
        • Dai B.
        • et al.
        Sanguinarine inhibits epithelial-mesenchymal transition via targeting HIF-1α/TGF-β feed-forward loop in hepatocellular carcinoma.
        Cell Death Dis. 2019; 10: 939
        • Keeley T.P.
        • Mann G.E.
        Defining physiological normoxia for improved translation of cell physiology to animal models and humans.
        Physiol Rev. 2019; 99: 161-234
        • Hectors S.J.
        • Wagner M.
        • Bane O.
        • Besa C.
        • Lewis S.
        • Remark R.
        • et al.
        Quantification of hepatocellular carcinoma heterogeneity with multiparametric magnetic resonance imaging.
        Sci Rep. 2017; 7: 2452
        • Patterson A.J.
        • Priest A.N.
        • Bowden D.J.
        • Wallace T.E.
        • Patterson I.
        • Graves M.J.
        • et al.
        Quantitative BOLD imaging at 3T: temporal changes in hepatocellular carcinoma and fibrosis following oxygen challenge.
        J Magn Reson Imaging. 2016; 44: 739-744
        • Bane O.
        • Besa C.
        • Wagner M.
        • Oesingmann N.
        • Zhu H.
        • Fiel M.I.
        • et al.
        Feasibility and reproducibility of BOLD and TOLD measurements in the liver with oxygen and carbogen gas challenge in healthy volunteers and patients with hepatocellular carcinoma.
        J Magn Reson Imaging. 2016; 43: 866-876
        • Park H.J.
        • Kim Y.K.
        • Min J.H.
        • Lee W.J.
        • Choi D.
        • Rhim H.
        Feasibility of blood oxygenation level-dependent MRI at 3T in the characterization of hepatic tumors.
        Abdom Imaging. 2014; 39: 142-152
        • Buffa F.M.
        • Harris A.L.
        • West C.M.
        • Miller C.J.
        Large meta-analysis of multiple cancers reveals a common, compact and highly prognostic hypoxia metagene.
        Br J Cancer. 2010; 102: 428-435
        • Ma L.
        • Hernandez M.O.
        • Zhao Y.
        • Mehta M.
        • Tran B.
        • Kelly M.
        • et al.
        Tumor cell biodiversity drives microenvironmental reprogramming in liver cancer.
        Cancer Cell. 2019; 36: 418-430.e416
        • Schito L.
        • Semenza G.L.
        Hypoxia-inducible factors: master regulators of cancer progression.
        Trends Cancer. 2016; 2: 758-770
        • Iommarini L.
        • Porcelli A.M.
        • Gasparre G.
        • Kurelac I.
        Non-canonical mechanisms regulating hypoxia-inducible factor 1 alpha in cancer.
        Front Oncol. 2017; 7: 286