Heavy LIFting: tumor promotion and radioresistance in NPC

Abstract

The epithelial-derived nasopharyngeal carcinoma (NPC) is a rare tumor in most of the world; however, it is common in southern China, northern Africa, and Alaska. NPC is often left undiagnosed and untreated until a late stage of disease. Furthermore, while radiation therapy is effective against this tumor, local recurrence due to radioresistance is an important clinical problem. In this issue, Liu et al. report on their identification of the IL-6 family cytokine leukemia inhibitory factor (LIF) as a serum predictor of local NPC recurrence following radiation therapy. The authors developed this initial finding to discover a role for the LIF/LIFR/mTORC1 signaling axis in NPC tumor cell growth as well as radioresistance.

NPC: a rare and difficult to diagnose epithelial cell cancer

Nasopharyngeal carcinoma (NPC) is a squamous-cell tumor that affects the epithelial cell lining of the nasopharynx. NPC is a rare tumor throughout the world, but it occurs with increased frequency in Southeast Asia and is tightly linked to EBV infection. While NPC can be cured by radiation therapy if diagnosed and treated early, often this cancer is not recognized until it has progressed to an advanced stage. Furthermore, approximately 20% of NPC patients have local recurrence following radiation (1). Indeed, a common cause of local recurrence and poor survival in NPC is radioresistance. While new imaging approaches have improved diagnosis and survival rates, new approaches to identifying biomarkers that predict local recurrence will be important in mitigating NPC disease burden and mortality.

Identification of serum biomarkers for NPC local recurrence

A minimally invasive approach to screening NPC patients would be to identify molecules secreted from the tumor environment into the blood that could be used as clinically predictive biomarkers. Therefore, Liu et al. screened the serum of NPC patients with local recurrence and compared it with serum from those who had gone into remission after radiation therapy (2). A panel of 20 cytokines was assayed, and a small group that included leukemia inhibitory factor (LIF), CXCL9, IL-10, IL-6, and SCF was among those differentially elevated in patients with local recurrence. Of the cytokines assayed, LIF was the most markedly different between NPC patients that responded to radiation therapy and those with local recurrence; therefore, LIF was further studied for its role in NPC pathogenesis and radioresistance.

Impressively, Liu and colleagues determined that LIF serum levels alone were predictive of NPC compared with healthy individuals (2). Furthermore, NPC patients with the highest levels of LIF were more likely to have local recurrence following radiation therapy; however, LIF levels were not predictive of either metastasis-free or overall survival. The authors also presented compelling immunohistochemical evidence that LIF levels in the tumor environment were higher than in normal tissues. Additionally, both LIF and the LIF receptor (LIFR) were overexpressed in NPC tumors as compared with adjacent tissue. These clinical observations suggested that LIF actually plays a role in NPC tumorigenesis rather than simply serving as a biomarker of local recurrence.

The LIF/LIFR interaction promotes NPC tumor growth and radioresistance

The authors next examined the role of LIF signaling in NPC tumor growth and radioresistance. LIF is a member of a family of IL-6–related cytokines that activate cell signaling pathways through both a unique and a common receptor, LIFR and glycoprotein 130 (gp130), respectively (3). LIF is known to activate the JAK/STAT3, PI3K, and ERK1/2 pathways to regulate cell growth. Recent evidence has implicated increased LIF signaling in a number of cancers including pancreatic (4), breast (5), glioblastoma (6), and thyroid (7) cancers. Therefore, strong rationale existed to investigate LIF as a growth-promoting factor in NPC tumorigenesis.

Liu et al. found that exogenous introduction of LIF signaled through LIFR to stimulate mTORC1-dependent phosphorylation of p70S6K and downstream STAT3 and ERK activation in NPC cells in culture (see Figure 9 of ref. 2). This signaling pathway was critical for NPC cell growth as well as growth of NPC xenografts in immune-deficient mice. Indeed, both soluble LIFR and rapamycin treatment prevented signaling through mTORC1 in these cells and inhibited tumor cell growth.

Follow-up immunohistochemical analysis in locally recurrent NPC tumor tissue indicated a strong upregulation and correlation of LIF, LIFR, and mTORC1 pathway activation markers. These data supported the notion that LIF may play a role in radioresistance. Indeed, the authors found that exogenous LIF could promote survival of NPC cell lines in culture and in xenograft studies following low-dose γ irradiation. Furthermore, LIF directly antagonized irradiation-mediated DNA damage response signaling, suggesting an acute mechanism in radioresistance.

Connecting EBV to LIF in NPC pathogenesis

It has long been appreciated that the ubiquitous human herpesvirus EBV is clonally present as a latent infection in nearly all NPC cases (8, 9). The virus expresses a restricted form of latency called latency II, in which the latent membrane proteins LMP1 and LMP2A are expressed along with one nuclear antigen EBNA1 and a set of viral noncoding RNAs (EBERs and BART miRNAs). The expression of LMP1 and LMP2A has been shown to constitutively activate the NF-κB (10) and PI3K pathways (11), among others, in epithelial cells and in NPC tissues (12, 13). In this study, the authors demonstrated that the viral LMP1 protein, through the NF-κB pathway, upregulated LIF mRNA and protein levels (see Figure 9 of ref. 2). Therefore, EBV infection and subsequent LMP1 expression in NPC tumor cells may explain the elevated LIF levels that contribute to disease pathogenesis and radioresistance.

Future questions and clinical opportunities

As expected in any important scientific study, Liu et al. (2) raise additional questions to be answered. These studies provide potential clinical opportunities for NPC, primarily in predicting recurrence following radiation therapy. The finding of increased LIF and LIFR expression in NPC tumors is compelling; however, it remains to be determined which cells in the tumor microenvironment produce the LIF that promotes tumor growth and radioresistance. Indeed, the authors provide evidence that ionizing radiation increases LIF levels, suggesting that stromal cells or immune cells within the tumor microenvironment are activated upon irradiation to produce LIF. Increased LIF in the microenvironment would then promote the growth and perhaps radioresistance of a population of tumor cells with increased LIFR levels. It is also possible, and likely, that the LIF/LIFR signaling pathway collaborates with other signaling pathways to drive tumor growth and radioresistance.

While the activity of LIF signaling through mTORC1 appears to be critical for NPC tumor growth, the collaboration between LIF/LIFR and the EBV latent membrane proteins will be important to assess in the future. Notably, LMP1 activation of the NF-κB pathway (10) and LMP2A-mediated activation of the PI3K/AKT pathway (11) may coordinate with the LIF/LIFR signaling pathway to promote tumor cell proliferation, survival, and possibly radioresistance. Indeed, recent studies suggest that the viral LMP1 protein can modulate the host DNA damage response (14, 15), which indicates collaboration between EBV and LIF for promoting radioresistance of NPC tumors following therapy.

An interesting, complementary component of LIF signaling that might be important for radioresistance is the Hippo/YAP pathway. This LIF/Yap axis has recently been implicated in mouse embryonic stem cell self renewal (16) and predicts a possible role for LIF signaling in cancer stem cell–like behavior (17). Because radioresistant cancer cells exhibit stem cell–like qualities (18), it is possible that LIF stimulation of Hippo/YAP may indeed contribute to NPC radioresistance. This exciting connection certainly warrants further investigation; it will provide insight not only into the mechanism of LIF-mediated radioresistance, but also probe the importance of stemness in the setting of NPC. Elucidation of this connection could ultimately contribute to new therapeutic targets aimed at preventing NPC recurrence.

Finally, this study addresses the question, What is a good serum biomarker for NPC recurrence and survival? Unfortunately, LIF levels alone were not predictive of overall or metastasis-free survival in NPC, despite correlating well with recurrence-free survival. Intriguingly, Liu et al. found a set of additional cytokines, including CXCL9, IL-10, IL-6, and SCF, that were also predictive of recurrence-free survival. Perhaps this cluster of cytokines together with LIF, EBV plasma viral load, or IgA levels (19) will have better predictive value for NPC survival and serve as a tool for identifying patients that fail to respond to radiation or chemotherapy. Identification of NPC prognosis-associated biomarkers remains a critically important area of research in the field. While imaging approaches to assess tumor recurrence have improved dramatically (1), a robust serological test to stratify or complement these approaches would provide great clinical benefit. Further analysis is warranted on larger numbers of patients in distinct NPC cohorts to address the potential value of LIF or related cytokines in predicting NPC progression.

In summary, the work of Liu et al. (2), along with other studies, has nicely highlighted how biomarker discovery can fuel the characterization of cell signaling pathways with potential for therapeutic relevance and mechanistic insight in tumor progression and resistance to standard therapy. This work will likely open the door for future studies focused on the role of the LIF signaling pathway in radioresistance of NPC and other tumors.