Understanding pathology in the context of physiological mechanisms: the practicality of a broken-normal view
- First Online:
- Cite this article as:
- Moghaddam-Taaheri, S. Biol Philos (2011) 26: 603. doi:10.1007/s10539-011-9258-2
- 108 Views
The topic of disease mechanisms is of clinical importance, as our understanding of such mechanisms plays an important role in how we approach devising treatments for disease. In this paper, I critique an argument made by Mauro Nervi, in which he asserts that pathology is often better viewed in the context of distinct theoretical mechanisms. I use this critique as a starting point to argue that viewing pathology as a broken-normal, malfunctioning mechanism is more therapeutically practical and more relevant to clinical drug design, than creating a theoretical separation of pathology from physiology.
KeywordsMechanismDiseasePhilosophy of medicineMalfunction
The topic of disease mechanisms is of clinical importance; a theoretical understanding of disease mechanisms plays a role in how we approach devising treatments to alleviate disease symptoms. In “Mechanisms, Malfunctions and Explanation in Medicine,” Mauro Nervi (2010) distinguishes between pathological and physiological mechanisms, claiming that pathological mechanisms are in many cases better viewed as distinct theoretical entities, to which malfunctions belong. I will briefly present Nervi’s argument for why pathological mechanisms should be viewed separately from physiological mechanisms, and I will critique Nervi’s argument that outcome variability, lack of range constraint, and ambivalence are properties that often necessitate a theoretical separation of pathological mechanisms from physiological ones. Then, I will use a critique of Nervi’s argument as a starting point to argue that understanding malfunction as a “broken” step within a physiological context is more practical and clinically relevant to drug discovery and treatment of disease than attributing malfunction to theoretically separate pathological mechanisms. I will endorse the view that broken-normal mechanisms may be a set of mechanisms that are malfunctions of a normal, physiological mechanism. Sketches of the overall structures of mechanisms serve to distinguish normal mechanisms from broken-normal ones and to show some of the ways in which a mechanism may be broken.
Pathological mechanisms as distinct theoretical entities
In arguing that pathological mechanisms are distinct theoretical entities, Nervi claims that no objective description of disease mechanisms exists, and thus disease mechanisms are described by researchers in the most pragmatic way (Nervi 2010, 217). He acknowledges that sometimes, physiological mechanisms can explain pathological phenomena as malfunctions of the normal, and roughly agrees that speaking of malfunction is often appropriate to the explanatory interest of clinicians and researchers (218). However, the bulk of his analysis focuses on cases where the pathological mechanism is not simply viewed as a malfunctioning normal mechanism (218). For such cases, Nervi claims that pathology is better viewed as theoretically separate from physiology (219). In arguing for the distinction of pathological mechanisms, Nervi contrasts pathological and physiological mechanisms, declaring that pathological mechanisms have properties of (1) outcome variability, (2) no range constraint, and (3) ambivalence, whereas physiological mechanisms have outcome invariability, a limited range of variation, and homogeneity (221). I will critique (1), (2), and (3) in the next section.
By outcome variability, Nervi claims that unlike physiological mechanisms, pathological mechanisms are not productive of regular mechanistic progressions, and vary with external and internal factors (221). Nervi claims that pathological mechanisms may result in a (2) wide array of outcomes, ranging from complete recovery to death, whereas physiological mechanisms are limited within what is considered to be a normal range of activity. Finally, Nervi argues that pathological mechanisms exhibit (3) ambivalence, meaning that they can potentially act adaptively or maladaptively depending on what regulatory factors are present (221). Nervi argues that the dissimilar properties of normal and pathological mechanisms indicates a theoretical distinction between them.
Problems in Nervi’s view and a critique of the theoretical distinction of pathology
It is of importance to address exactly what is meant by the term “mechanism.” Nervi accepts the Machamer, Darden, and Craver (MDC) characterization of mechanism, which asserts that “mechanisms are entities and activities organized such that they are productive of regular changes from start or set-up to finish or termination conditions” (Machamer et al. 2000, 3). Although Nervi employs terminology of “elements and activities” (Nervi 2010, 219) similar to MDC terminology, Nervi’s argument for the properties of pathology that often necessitate a distinction of pathological from physiological mechanisms is problematic, given his acceptance of the MDC view of mechanisms. According to MDC, each mechanism produces a single phenomenon—not, as Nervi’s property (2) demands, a range of phenomena. Following from the MDC view, each individual pathological mechanism will produce a single, regular pathological phenomenon, and each unique phenomenon produced can be attributed to the set-up conditions and/or the activities occurring along the mechanistic pathway. Hence, according to the MDC view, although a mechanism may be pathological, it will still produce regular changes that are dependent on the conditions under which the earlier steps in the pathway occurred. If the MDC view is fully adopted by Nervi, then it is not correct that a single pathological mechanism will produce, in of itself, a range of phenomena. Rather the mechanism is defined in the context of its internal and external environment, which may define the set-up and/or termination conditions. However, Nervi is neither consistent with the MDC view he relates to, nor does he provide a way to discredit the validity of MDC in providing a mechanistic explanation of pathological mechanisms. If Nervi is to argue that a pathological mechanism does not yield regular changes, given an exact account of set-up conditions and of participating entities and activities, more explanation is needed for how to deviate from MDC.
The properties that Nervi takes as evidence for theoretically separate pathological mechanisms are arguably more effective when explained within a physiological context. Nervi’s property (1) outcome variability is not sufficient to distinguish pathological mechanisms from physiological ones, because outcome variability can be discussed in a physiological context. Namely, one could assert that the branching pattern of outcomes observed in pathology results from malfunction of one or more stages within a linear physiological mechanism.1 Malfunction may be attributed to the stage at which the pathological mechanism deviates from the physiological one, such that it interferes with an organism’s ability to survive and thrive. Importantly, I will appeal to the MDC notion of function, which asserts that “functions are the roles played by entities and activities in a mechanism” (Machamer et al. 2000, 6). One may infer that MDC would therefore claim that malfunctions occur when the roles played by the entities and activities of a physiological mechanism are changed.2
If the proto-oncogene encounters a point mutation under the same set-up conditions, it will produce a regularly operating, malfunctioning mechanism of cell proliferation. The physiological mechanism by which the proto-oncogene regulates cell proliferation has been interrupted, thus leading to cancer (Fig. 2b). However, if the proto-oncogene entity does not become mutated, and only one copy of the tumor-suppressor gene becomes mutated, then the pathway will continue to yield physiological outcomes (Fig. 2c). If both copies of the tumor-suppressor gene become mutated while the proto-oncogene remains undamaged, the mechanism malfunctions at the stages involving the tumor-suppressor gene to yield a cancerous outcome (Fig. 2d). Hence, the progression of the pathological process is contingent upon which entities and activities of a mechanism become damaged and thus malfunction. Each unique malfunction yields a regularly operating mechanistic pathway and should be considered a different kind of malfunction of the physiological mechanism.
Nervi’s condition (2) a lack of range constraint is also not sufficient to theoretically separate pathological mechanisms from physiological ones. Namely, depending on which stage malfunctions within a physiological mechanism, the severity of associated symptoms could increase or decrease. I argue that “range of variability” should be replaced with the notion that if unique set-up conditions or abnormal conditions along a mechanistic pathway are present, then unique, regularly operating, broken-normal mechanisms will arise. In other words, pathological mechanisms should not be viewed to have a range of variability, and should not be viewed as distinct theoretical entities. Rather, unique set-up or mechanistic conditions alter the normal physiological pathway (Fig. 1). Abnormal conditions entail having malfunctioning entities and/or activities. That is, a mechanism explains how a phenomenon may have been produced (MDC 2000, 3). Thus a physiological mechanism that includes a malfunction can explain how the phenomenon of disease phenotype was produced. Figure 1 demonstrates the broken-normal view.
If the mechanism is not understood, Nervi may be justified in utilizing a sketch that distinguishes pathology from physiology. In that case, from a clinical perspective, knowledge of the endpoint symptom of the pathology may be enough to treat that symptom. However, if the mechanism or substantial parts of the mechanism are understood, then such a sketch cannot adequately account for the point at which a mechanism malfunctions, and thus, the sketch is not useful in restoring the original mechanism by targeting the malfunctioning stage. Additionally, it is unreasonable to claim that one malfunction, in the context of the same background conditions can yield multiple outcomes. Namely, if all entities and activities remain constant, then a unique malfunction presented to the same background conditions will yield a regular outcome. What Nervi believes to be a range of outcomes exhibited by pathological mechanisms is due to an incomplete knowledge of the details pertaining to the malfunctions of the physiological mechanism that lead to disease.
The ambivalence that Nervi attributes as property (3) to pathological mechanisms can also be viewed within the context of malfunction of a physiological mechanism. Specifically, an adaptive physiological mechanism, such as an immune response, may malfunction at a given stage, as shown above, along the mechanistic pathway, thus resulting in maladaptive autoimmune disease. Maladaptive properties may thus be viewed as malfunctions of physiological, adaptive properties. Hence, the maladaptive properties that Nervi describes do not adequately justify the need to view pathological mechanisms as separate from physiological mechanisms. It is important for Nervi to adequately distinguish between physiological function and adaptive function. A particular pathology may confer a selective advantage to an organism, while still being a broken-normal mechanism nonetheless. For example, sickle cell anemia may afford the afflicted individual the ability to survive and thrive, but only in an environment that carries malaria. Thus, the adaptive function of sickle cell anemia creates a mechanism, with entities and activities due to the malaria-carrying environment. The fact that malaria resistance is conferred by the malfunction leading to sickled blood cells does not change the fact that sickled blood cells are a malfunction of the normal. Hence, the property of ambivalence—although an interesting property—is not enough to theoretically remove pathology from a physiological context.
Practicality of a broken-normal view
Embracing a broken-normal view of disease mechanisms is more practical to drug design and therapy than adopting a view that separates pathological and physiological mechanisms. A broken-normal is a malfunction that occurs at a given stage or number of stages within the pathway of a physiological mechanism. Hence, the pathway of disease can be adequately described as malfunctions within a physiological mechanistic pathway, when there is knowledge of the mechanism at hand. According to Nervi’s discussion of malfunction (218), he acknowledges that in many cases, talk of malfunctioning mechanisms is practical. It is reasonable that if the end symptom of a disease is known and can be treated, while the mechanism of the disease is not understood, that for therapeutic purposes, understanding pathology as a separate theoretical entity may be useful for treatment. Thus, there may be a short-term therapeutic advantage in establishing a theoretical separation of pathology from physiology. However, embracing such a therapeutic perspective is not valuable in the long-term, as a more complete understanding of malfunction within a physiological context is what drives the production of powerful drug targets and therapies.
An example of the value of the broken-normal view lies in cystic fibrosis research. In cystic fibrosis, cystic fibrosis transmembrane regulator (CFTR) protein does not function properly or is absent (Silverthorn 2010, 165), and cannot mediate ion transport. The Class II CFTR mutation is the most common. In Class II mutants, the CFTR channel is misfolded and is thus degraded before it is able to reach the apical membrane of the epithelial cell (Rowe et al. 2005, 1998). One study currently “raise[s] the possibility that rescue of misfolding diseases may be achievable through understanding the cargo-specific roles of chaperome components…” (Wang et al. 2006, 804). The researchers’ approach delineates the importance of a physiological context for disease. Practical treatment of Class II mutation involves correcting malfunction (misfolding of CFTR) in one stage of the physiological pathway, which therefore corrects stages downstream of the malfunction. The broken-normal view is thus more appropriate than Nervi’s view when the goal is to treat points throughout the pathway, not just the endpoint, because there is knowledge of the mechanism.
Class I mutations of CFTR occur as premature truncations of the gene or by the presence of a nonsense allele (Rowe et al. 2005, 1998). Experiments aimed at restoring Class I mutations have used agents that deter premature stop codons, thus correcting the gene sequence for CFTR (Wilschanski et al. 2003, 1433). Although such experimental therapies have not been clinically successful, they target a specific malfunction along the physiological pathway, and restore normal physiology at the malfunctioning stage. In Classes I and II, the CFTR gene is mutated in two different ways, thus producing two different malfunctioning gene products. If we viewed the defective CFTR as a separate pathological mechanism, we could not speak efficiently of “rescuing” or “correcting” malfunctioning steps of a pathway back to physiological.
Nervi is vague in his discussion of clinical pragmatic goals, thus I will specify such goals. Namely, if we separate pathology into discrete theoretical mechanisms, it becomes difficult to envision how to target specific stages within the mechanism for therapeutic intervention, in order to restore normal physiology. Separating pathology from physiology distances the notion of disease from the practical issue of therapy. Likewise, it is not efficient to attempt characterizing every possible outcome of a malfunction as a separate pathological mechanism, as there are so many possibilities. Rather, by acknowledging that malfunction within entities and activities of identified stages of physiological mechanisms can lead to regular, pathological phenomena that are unique to each malfunctioning stage, we can efficiently characterize the malfunctioning stage, and identify it as the stage for therapeutic intervention required for restoring physiological function.
I have criticized Nervi’s argument that pathological mechanisms should in many cases be distinguished from physiological mechanisms, by arguing that the properties he takes to be indicative of discrete pathological mechanisms can in fact be better understood in the context of physiological mechanisms. I have argued that a broken-normal view within the context of physiological mechanism is more clinically relevant under conditions in which the mechanism of disease is known. In doing so, I have acknowledged that Nervi does not completely deny the importance of malfunction, but rather that he places too much emphasis on a separation of pathological and physiological mechanisms in contexts that are not necessary. Nervi’s view is valid when little is known about the mechanism while the end-symptoms of disease are defined and can be treated. In contexts in which the mechanism is known and malfunctions can be targeted, a broken-normal view is more practical for designing therapies.
I thank Tudor Baetu for this point, made in a discussion of Nervi’s argument.
I wish to thank Lindley Darden for many helpful comments on earlier drafts, as well as the Maryland Philosophy of Biology Reading Group, including Lane DesAutels, Tudor Baetu, and Erika Milam for helpful discussion of Nervi’s paper.