Cancer, cachexia, prostanoids, and NSAIDs

It is a major medical challenge to provide treatments that can help patients with advanced cancer – to relieve pain, remedy functional impairment, and counteract wasting and fatigue. Cachexia, the crucial feature of which is muscle wasting, with or without loss of fat, and with associated weakness, disability, and reduced quality of life, affects a majority of the patients with advanced cancer and has been estimated to be the direct cause of death in a substantial proportion of the cases. Unfortunately, therapeutic options are limited. Progestins may increase appetite and body weight, and they, to some extent also glucocorticoids, are in clinical use, but they do not counteract loss of muscle mass [1]. New approaches are needed. While the pathophysiology of cancer cachexia is not fully understood, research over the past two decades has yielded insights into some of the underlying mechanisms [2,3]. This research has, in particular, established the role of systemic inflammation in cachexia and the involvement of cytokines and other pro-inflammatory mediators, identified certain tumor-derived cachexia-inducing factors, and characterized the resultant metabolic dysregulation which leads to proteolysis and lipolysis and thereby loss of skeletal muscle mass and fat depletion. This knowledge may bring clues to better therapies. However, while numerous strategies have been attempted in cancer cachexia and many are currently being investigated [1,4], no satisfactory treatment has so far been established. Therefore, both more basic research and clinical trials are needed to develop new strategies and to exploit the information that already exists.

One therapeutic principle that has been attempted in patients with advanced cancer and cachexia is the use of non-steroid anti-inflammatory drugs (NSAIDs). These agents are inhibitors of the cyclooxygenases (COX-1 and COX-2), i.e. the enzymes that are rate-determining in the synthesis of prostaglandins and other prostanoids from arachidonic acid [5]. The prostaglandins mediate a large number of cellular effects involved in many physiological regulations and pathologic conditions [6]. A notable role of prostaglandins is their important participation in various types of inflammation. Activation of cyclooxygenases, particularly the inducible form, COX-2, is triggered by cell damage, various pro-inflammatory cytokines, and possibly tumor-derived factors. This increases the levels of prostaglandins, which in turn act in concert with cytokines, thus participating in fundamental inflammation-promoting synergistic loops and networks. Consequently, inhibition of cyclooxygenases is an important and well-established pharmacological mechanism exploited in the treatment of inflammatory disease. Furthermore, with the emerging realization that inflammation is a crucial factor in the pathophysiology of cancer cachexia [2,3], it has been logical to try NSAIDs in this condition.

In this issue of Acta Oncologica, Solheim and colleagues [7] have reviewed clinical studies of the effects of NSAIDs in cancer cachexia. They find that, despite great heterogeneity in terms of study design, number of patients, type of cancer, clinical parameters, and definition of effect criteria, the studies that were reviewed, taken together, support the notion that NSAIDs can have beneficial effects in cachectic cancer patients. From a large number of retrieved articles, 13 studies were included in the review. Due to the heterogeneity of the studies, a formal meta-analysis was not appropriate, and Solheim et al. made a qualitative analysis and synthesis of the data contained in these papers. They conclude that the reviewed studies suggest that NSAID treatment in patients with cancer cachexia increases body weight, with a median of around 2.5 kg after six weeks and about 5 kg after 12 weeks, and they find some evidence of improved physical performance and self-reported quality of life. Basically the same conclusion is drawn in another very recent review where Reid et al. [8], restricting their analysis to four randomized and controlled studies, found evidence in support of a positive effect of NSAIDs, in terms of weight gain, performance status, survival, and quality of life. It is also worth noting that both these reviews find that side effects of NSAID use were not prominent in the studies that were evaluated.

However, both Solheim et al. [7] and Reid et al. [8] conclude that the existing evidence is insufficient for recommending widespread use of NSAIDs in cachectic patients. This seems reasonable, in view of the enormous heterogeneity of the studies and the frailty of many of the individual studies. On the other hand, some of the observations made in these studies are quite striking and have brought up several issues that will be important in further research and reflection in this field. A notable example is the pioneering work by Lundholm et al., including the finding of a marked increase in survival when patients with advanced cancer were treated with indomethacin [9]. Further clinical investigations, with controlled and sufficiently large studies to establish more firmly the effects NSAIDs on cachexia and survival, are highly desirable, as also pointed out by Solheim et al. The recently published consensus paper on definitions and classification of cachexia [10] should be of value in designing clinical trials in this field. However, at the same time more studies of the mechanisms underlying the effects of NSAIDs in cachexia may provide new clues for further improvement of the therapy. More generally, great efforts are currently being invested to develop novel therapies directed at specific signaling mechanisms in cancer [11]. In other approaches to cancer cachexia, penetrating molecular investigations and animal studies have led to promising results, such as the demonstration that pharmacological blockade of ActRIIB (the receptor for activin and myostatin) leads to reversal of muscle wasting and dramatically increases survival in tumor-bearing mice [12]. Hopefully, further mechanistic studies may also provide progress in understanding the role of cyclooxygenases and the possible therapeutic benefit of NSAIDs in cancer cachexia.

It is not known to which extent the effects of NSAIDs in cachexia reflect general anti-tumor actions. Solheim et al. argue that the best way to control cancer cachexia is to control cancer progression. It is not difficult to agree with that. Although the possibility exists that NSAIDs may have direct actions on the wasting tissues or on homeostatic mechanisms, it is likely that they exert at least a major part of their effects via the tumor, by disrupting cancer-induced, prostaglandin-mediated, pro-cachectic signals. A large body of evidence implicates cyclooxygenases, particularly COX-2, and prostaglandins in the development and progression of malignant tumors [13,14]. Thus, COX-2 is commonly upregulated in cancers, prostaglandins can exert a number of tumor-promoting effects in vitro, and NSAIDs have been shown to inhibit tumor growth and metastases in animal experiments. In humans, the early, but disputed, data suggesting a protective effect of NSAIDs on the risk of developing colorectal cancer have been confirmed by recent investigations [15]. Regular use of aspirin (acetylsalicylic acid) may also reduce the risk of several other cancers [16]. Furthermore, evidence that inhibition of cyclooxygenases can interfere with the progression of human cancer has been obtained in prospective studies showing that treatment with aspirin after the diagnosis of colorectal cancer is associated with a lower risk of colorectal cancer-specific (and overall) mortality, particularly in individuals with tumors overexpressing COX-2 [17,18]. Very recently a new level of specificity has been reached as it has been shown that this inhibitory effect of aspirin on cancer progression is confined entirely to those patients whose tumors harbor activating mutations in phosphoinositide 3-kinase (PI3K) [18].

For the effects of NSAIDs on cancer cachexia, we do not have any similar molecular data. Most likely COX-2 mediates the inflammatory reactions and probably also the cachectic effect [19]. Several prostaglandins and thromboxans are formed downstream of the cyclooxygenases, and they act via at least 10 different receptors [20]. It is meaningless to speculate on the role of distinct prostaglandins and receptors in cachexia until the gross effects of NSAIDs are more firmly established. However, agents that inhibit specific prostanoid receptors are currently being developed and may represent better and more selective therapeutic alternatives in the future.

The review by Solheim et al. [7], like that of Reid et al. [8], shows that NSAIDs may have beneficial effects in cancer cachexia, but also that this has not been sufficiently documented and therefore the treatment can presently not be recommended outside clinical trials. Hopefully adequate studies, including trials with NSAIDs in combination with other agents, will be performed. It is hard to avoid noting the striking contrast between the enormous efforts and resources spent on developing novel and very expensive drugs targeted at signaling pathways, many of which have turned out to have only marginal effects, and the lack of good studies trying out the effects of relatively cheap drugs in patients with very advanced cancer.

Declaration of interest: The author reports no conflicts of interest.