Oral mucosal injury in oncology patients: perspectives on maturation of a field
Department of Oral Health and Diagnostic Sciences, School of Dental Medicine, University of Connecticut Health Center, Farmington, CT, USA
Program in Head & Neck Cancer and Oral Oncology, Neag Comprehensive Cancer Center, University of Connecticut Health Center, Farmington, CT, USA
Correspondence: Douglas E. Peterson, DMD, PhD, FDS RCSEd, Department of Oral Health and Diagnostic Sciences, School of Dental Medicine, University of Connecticut Health Center, 263 Farmington Avenue, Farmington, CT 06030-1605, USA. Tel: 860 679 3421, Fax: 860 679 4760, E-mail: email@example.com
This work was funded in part by NIH/NIDCR 1R01DE021578-01.
In the past decade, there have been important strategic advances relative to pathobiological modeling as well as clinical management for oral mucositis caused by cancer therapies. Prior to the 1990s, research in this field was conducted by a relatively small number of basic and clinical investigators. Increasing interest among researchers and clinicians over the past twenty years has produced a synergistic outcome characterized by a number of key dynamics, including novel discovery models for pathobiology, increased experience in designing and conducting clinical trials, and creation of international collaborations among cancer care professionals who in turn have modeled clinical care paradigms based on state-of-the-science evidence. This maturation of the science and its clinical translation has positioned investigators and oncology providers to further accelerate both the foundational research and the clinical modeling for patient management in the years ahead. The stage is now set to further capitalize upon optimizing the interactions across this interface, with the goal of strategically enhancing management of patients with cancer at risk for this toxicity while reducing the cost of cancer care.
Oral mucositis in patients with cancer represents a complex toxicity with multivariate etiology (Al-Dasooqi et al,2013; Miller et al, 2012; Peterson et al, 2012; Al-Dasooqi et al, 2013; Cheng et al, 2013) (Figure 1). The lesion is directly caused by high-dose cancer therapy such as induction chemotherapy and/or head and neck radiation. Despite this delineation of high-risk groups, however, there are many questions at the basic as well as clinical level that if answered could transform the clinical management of these patients.
Of additional, recent interest is the unique clinical expression of mucosal injury that has emerged over the past 5 years in association with molecularly targeted therapies such as mammalian target of rapamycin (mTOR) inhibitors (Pilotte et al, 2011; Elting et al, 2013) (Figure 2). As with oral mucositis in patients receiving conventional cancer therapies, oral mucosal lesions associated with novel biologics for cancer treatment can have significant clinic consequence. Further investigation of this lesion is warranted, including potential development of a predictive biomarker to predetermine patient's response to these molecularly targeted agents (Reynolds et al, 2013).
Oral mucositis can adversely impact the cancer patient's clinical course including (i) severe oral pain that compromises enteral intake and oral hygiene care, (ii) need for hospitalization and opioid analgesics, and (iii) elevation in risk for infection (Sonis, 2009). In patients with neutropenic cancer, the ulcerative oral lesions can be a portal of entry for potentially fatal systemic infection. Severe mucositis can also cause interruptions and/or dose reductions in cancer treatment protocols that can in turn lead to decreased patient survival (Peterson and Lalla, 2010). Given the importance of these clinical complications, there have fortunately been strategic scientific and clinical advances in recent years that are producing changes in the paradigm from historically based symptom management to targeted prevention and treatment of the condition.
Observations from the clinical sector: valuable clues
Frequency and severity of oral mucositis have classically been viewed as governed primarily by type, mechanism, and delivery sites of the cancer therapeutic.
For example, high-dose head and neck radiation typically causes mild injury in most patients when the oral mucosal tissue is exposed to at least 1500–2000 cGy; moderate-to-severe mucositis then predictably occurs as the radiation dose approaches and exceeds 5000 cGy. Intensity-modulated radiation therapy (IMRT), although associated with higher doses to selected non-target tissue vs static beam techniques, does not appear to cause increased acute mucosal toxicity (Kruser et al, 2013). High-dose chemotherapy can cause oral mucositis ranging in severity from mild to severe, as is the case with gastrointestinal mucositis as well. Oral mucositis occurs in up to 40% of patients receiving standard dose chemotherapy (Peterson et al, 2012). By comparison, high-dose conditioning chemotherapy in the hematopoietic stem cell transplant setting can cause oral mucositis in approximately 80% of patients. Intestinal mucosal injury in contrast is characterized by abdominal pain and diarrhea, the latter of which can occur in approximately 40% of patients. Interestingly, tissue at select other sites such as conjunctival and vaginal mucosa is seemingly ‘privileged’ and does not exhibit injury secondary to cancer therapy in most patients. Reasons for this latter phenomenon are not clear and warrant further study.
Quality of life can be significantly compromised by this constellation of oral and gastrointestinal toxicity. For example, Elting and coworkers have confirmed strong associations between mucositis and adverse outcomes at the clinical as well as patient-reported level (Elting et al, 2008). Although supportive care and palliative measures can reduce the discomfort of mucositis, there continues be a significant unmet medical need regarding prevention and treatment.
The recently documented mucosal toxicities associated with molecularly targeted cancer agents are creating new interest at the laboratory and clinical level as well. Pathobiology of these latter oral lesions is not well delineated (Al-Dasooqi et al, 2013). Insights into new research pursuits may be fostered by the observation that the clinical trajectory with these lesions is distinctly different from mucositis observed in patients receiving treatment with conventional cancer agents. Time of onset, physical appearance, and an often concurrent dermal rash in patients receiving targeted cancer agents are but some of the unique features of this recently emergent oral mucosal lesion.
Evolution of the field
Lessons from the clinical sector as described above have informed new directions for research that could ultimately transform clinical management of this toxicity (Table1). Historically, oral and gastrointestinal mucositis has been viewed by many cancer care clinicians as an inevitable consequence of high-dose chemotherapy or radiation. Over the past twenty years, however, a substantive change in this paradigm has occurred for reasons including novel discovery models directed to pathobiology, increased experience in designing and conducting clinical trials, and creation of international collaborations among cancer care professionals who in turn have modeled clinical care guidelines based on state-of-the-science evidence.
Table 1. Oral mucositis: evolution of the field
Selected key developments
These organizations have adopted/adapted or refer to the MASCC/ISOO mucositis guidelines.
Prior to and during the 1970's
Supportive oral care based on clinical experience
Initiation of research studies into pathobiology and clinical trials
Creation of international health professional organizations directed to mission of supportive care in cancer
Continued expansion of basic and clinical research, including pathobiology
Alliance of two supportive care organizations:
Multinatl Assoc Supp Care Cancer & Intl Soc Oral Oncol (MASCC/ISOO)
First two Am Soc Clin Oncol (ASCO) Educational Sessions addressing mucositis (2003 and 2004)
First evidence-based mucositis guidelines published by MASCC/ISOO (2004), followed by “Mucositis” incorporated as MeSH term by the National Library of Medicine as a direct outcome (2006)
Updated evidence-based mucositis guidelines published by MASCC/ISOO (2007)
First MASCC/ISOO Mucositis Research Workshop (2007)
Mucositis guidelines published by other organizations (2007-2010):
National Cancer Institute Physician Data Query (PDQ) website comprehensively updated: Oral complications of chemotherapy/head & neck radiation (2011)
Third ASCO Educational Session on mucositis (2012)
Version 3 of the Clinical Practice Guidelines published by MASCC/ISOO (2012 and 2013)
First-in-kind Gordon Research Conference “ Mucosal Health & Disease” (2013)
Sixth MASCC/ISOO Mucositis Research Workshop (2013)
These scientific and clinical advances have set the stage for the development of molecularly targeted mucositis therapeutics. The first of these biologics (palifermin) (Sonis, 2010a) was approved in December 2004 by the United States Food and Drug Administration (FDA) for the prevention of oral mucositis in patients with hematologic malignancy and who were undergoing a high-dose chemotherapy conditioning regimen followed by hematopoietic cell transplantation (HCT). Given its pleiotropic mechanisms across epithelium and submucosa, palifermin has served as an important first-in-class prototype for the modeling of mucositis therapeutics in relation to key network pathways. In the years since 2004, there have been a number of pharmacologic and biological agents that have moved through the phase 1, phase 2, and phase 3 pipeline (Keefe et al, 2008; Peterson et al, 2012). (Table2). Unfortunately, however, there has not been a new approval by the US FDA for such pharmacologics or biologics since the palifermin approval in 2004.
Table 2. Selected agents in development for the prevention of/treatment for mucositis
Stage of trial vs cancer treatment
Targeted cancer therapy
Source: ClinicalTrials.gov, industry Web sites and PubMed.
Modified from: Peterson DE, Keefe DM, Sonis ST (2012).
Adenoviral vector-mediated transfer of keratinocyte growth factor cDNA
Lactobacillus CD 2 lozenges
Light-emitting diode (LED) therapy
Chamomilla recutita mouthwash
IZN-6N4 (botanical extracts)
Light-emitting diode (LED) therapy
Low-level laser light therapy
Doxepin hydrochloride rinse
Impact® Enteral Nutrition
Impact® Enteral Nutrition
There have, however, been numerous and important advances in science and clinical development in addition to industry-based initiatives. For example, novel strategies currently in preclinical study at the National Institutes of Health include Tempol for the prevention of oral mucositis (Cotrim et al, 2012) as well as gene therapy for the treatment for chemoradiation-induced oral mucositis (Zheng et al, 2009). Research by other investigators has demonstrated that administration of recombinant human epidermal growth factor vs placebo in HCT patients resulted in improved secondary endpoints in patients experiencing World Health Organization grade ≥3 oral mucositis, including reduced duration of hospitalization (Kim et al, 2013). In addition, novel tissue engineering models of oral mucosa (Colley et al, 2013) hold promise for strategically advancing preclinical development of these technologies. This collective scope of science is in turn being further advanced by novel modeling directed to prediction of the mucositis trajectory in patients, including but not limited to genetic (Garg et al, 2012) and imaging (Calantog et al, 2013) strategies.
These and related research pursuits will hopefully create future opportunity for clinicians to customize molecularly based management, based on a priori risk assessment for a specific patient for whom cancer treatment is being planned.
Current molecular paradigm
The molecular model for mucositis pathobiology continues to mature at a rapid pace. The historically based component of clonogenic progenitor stem cell death of oral basal epithelium and/or intestinal villi continues to be a key constituent of the contemporary model. What has changed over the past 15 years, however, is that mucositis is now viewed as a complex result of a highly interdependent series of molecular, cellular, and tissue changes affecting both epithelium and submucosa (Sonis, 2009, 2010b, 2011; Bowen et al, 2011; Mougeot et al, 2011; Chen et al, 2011; Peterson et al, 2012; Logan et al, 2007, 2009; Al-Dasooqi et al, 2013; Yasuda et al, 2013) (Figure 3). These alterations can in turn be further influenced by external factors unique to the specific oral or intestinal environment (Al-Dasooqi et al, 2013; Stringer, 2013) and for which state-of-the-science technology can be utilized for research (Diaz et al, 2012). This evolving increasingly complex pathobiology model provides an essential conceptual framework for the development of mucositis therapeutics, including drugs, biologics, and devices.
Advances in genomics in recent years have specifically provided new strategic technologies to pursue the molecular complexity of mucositis (Sonis et al, 2007; Hahn et al, 2010; Al-Dasooqi and Keefe, 2011, 2013Al-Dasooqi et al,2011; Mougeot et al, 2011, 2013). While this complexity represents a challenging barrier, it also defines well the opportunity for research directed to prediction of severity of mucositis in an individual patient as well as prediction of response by that patient to a mucositis therapeutic intervention. The genomic technology currently being utilized holds high promise for identifying answers to these important new frontiers.
As previously noted, the advent of molecularly targeted cancer treatments in clinical practice has produced new types and trajectories of toxicities, including mucosal lesions (Pilotte et al, 2011). The recurrent aphthous ulcer-like lesions associated with selected mTOR inhibitors have become a prototypic example (Figure 2). This condition occurs in approximately 50% of patients receiving this class of agents and is characterized by discrete aphthous-like lesions in the days and weeks following administration of treatment. It can be sufficiently painful as to cause dose reduction in the cancer treatment. Pathobiology is not well defined; given the unique presentation and trajectory, one can logically hypothesize that there are unique mechanistic features to the lesions as well. Early clinical experience with topical, intralesional, and/or systemic steroids has produced promising results for treatment and suggests as well a fundamentally different pathobiology from that of oral mucositis in patients receiving chemotherapy or radiation. Additional research is needed.
Novel investigations utilizing systems biology technology
Use of systems biology is becoming more pronounced in fundamental research as well as in application to systems of clinical relevance (Michelson et al, 2006; Molina et al, 2010; Sonis et al, 2013). If properly leveraged, systems biology has a clear role to contribute in the area of oral disease, including oral mucositis. In this context, ‘systems biology’ is being utilized to delineate phenomena at the molecular, cellular, and tissue levels.
The underlying concept of systems biology is that the whole is greater than the sum of the parts. Whereas one can study the individual components of a biological pathway (e.g., components of the inflammation pathway), understanding what each molecular species does will only inform part of the story (Vodovotz et al, 2008). Interactions among these species are critical and provide a new dimension of information that would otherwise be missed. As a result, a truly comprehensive understanding can be achieved only by studying the system as a whole. It becomes possible with this level of understanding to develop and ultimately intervene with more effective clinical strategies for disease treatment.
The system of molecular interactions being analyzed can often be highly nonlinear and convoluted in nature. For example, it may not be clear what impact up-regulation or down-regulation of a given gene might have on the system as a whole. Perhaps more perplexingly, it may seem obvious what the effect of genetic perturbation should be, but in reality the opposite effect is observed. To deal with these non-intuitive phenomena, it is possible to develop computational models of biomolecular behavior (Thiele et al, 2013). Although significant advances in modeling cellular phenomena at the whole-cell level have been achieved (Karr et al, 2012, 2013), considerable work remains. It is now possible to take advantage of computational models to organize, understand, and, to some level, predict physiological behavior to complex perturbations (Kreeger and Lauffenburger, 2010).
Models based on mechanistic knowledge of biomolecular phenomena tend to provide the most insight into the biological system of interest. However, oftentimes the mechanism may only be partially understood or may not be known at all. In such cases, machine learning can provide a powerful complementary strategy to model development. ‘Machine learning’ refers to the development of computational systems that learn about a process, whether biological or otherwise, from data in order to ultimately make predictions regarding that process (Tarca et al, 2007; Yip et al, 2013). For example, recent work using machine learning has been used for the prediction of risk of oral mucositis in stem cell transplant recipients as well as to predict the response of individuals to SCV-07 (gamma-d-glutamyl-l-tryptophan) for the attenuation of chemoradiation-induced oral mucositis (Alterovitz et al, 2011).
In general, the ability to mathematically model biological systems provides a whole host of advantages not otherwise realizable. New approaches for conducting therapeutic optimization (Swierniak et al, 2009; Rini, 2010) as well as drug discovery and development (Yang et al, 2007) become possible.
The basis for therapeutic optimization arises from the fact that once a mathematical model is available to describe the system of interest, it is possible to use optimization theory to determine the most efficacious treatment strategy. Oral mucosal injury as a secondary toxicity of head and neck cancer treatment provides an excellent example. If a comprehensive high-quality quantitative systems biology model describing oral mucositis were to become available, it would be possible to determine the amount of chemotherapy or radiation therapy required that would minimize occurrence of secondary toxicities while maintaining cancer therapeutic efficacy.
Additionally, with a comprehensive mathematical model of a clinically relevant biomolecular network, it would be possible to identify which specific pathways should be modulated in order to inhibit disease effects (Pawson and Warner, 2007; Deisboeck et al, 2009). These pathways could then become targets against which to design new drugs as well. Identification of these pathways may be accomplished with relatively simple mathematical approaches, such as sensitivity analysis. Sensitivity analysis provides insights into which pathways are most robust and which are most sensitive to any perturbation that might be introduced. This approach permits evaluation of the system-wide impact or lack thereof of a perturbation to a given pathway in a network. Given the level of redundancy found in biological systems, such information would have particular value in ruling out certain classes of drugs. For example, a drug that down-regulates a given pathway might be identified, only to find that some redundant mechanism compensates for the drug effect. Identifying this possibility computationally could prove extremely valuable relative to saving research time and cost.
By inverting the described approach, it is also possible to determine the mechanism of action of a drug. For example, by collecting data regarding physiological responses to a drug, the computational model of the affected system may be modulated by turning on or turning off various pathways until the computer simulates the same physiological response. By reviewing which pathways are being modulated, it is possible to develop a hypothesis of the mechanism of action for the drug.
In sum and as delineated later in this paper ('Key research questions'), the interface of the biological with the computational sciences promises to create new insights and opportunities at basic as well as clinical research levels. Capitalizing upon systems biology technology represents an important emerging frontier that could contribute in the years ahead to strategic research advances relative to oral mucositis.
Translation into clinical practice guidelines
The many important advances in molecular modeling of oral mucositis in recent years continue to define new opportunities for management at the clinical level. Evidence-based guidelines for high-quality supportive care of cancer patients are becoming increasingly valuable in clinical practice, including prevention of and treatment for oral complications.
Emergence of evidence-based mucositis guidelines in 2004 represented a key first step in the maturation of the field (Rubenstein et al, 2004). In the years following this first set of evidence-based mucositis guidelines, additional versions have been developed, ranging from expert opinion through systematic reviews to meta-analyses (Keefe et al, 2007; Peterson et al, 2011, 2012; Lalla, 2013) (Table1).
The literature delineating the importance of oral management of oncology patients receiving intensive cancer therapies is broad and deep (Natl Cancer Inst PDQ®, 2013). In many of these patients, precancer therapy oral care can substantially reduce clinical risk of acute and/or long-term adverse sequelae, including prolonged hospitalization, extensive supportive care management, and cancer treatment dose reduction and/or delay. Given the clinical and economic value of preventive oral management throughout the cancer care continuum, it continues to be essential that state-of-the-science guidelines be developed and implemented in ongoing fashion.
Having said this, the practical reality is that production of these guidelines may not uniformly translate effectively into clinical practice. This reality at present is principally due to limitations in steps 3 and 4 below in relation to the several key steps associated with the creation of oncology clinical practice guidelines:
Production, including quality of the literature search, weighting of evidence, and incorporating interprofessional input;
Dissemination, including clinical user-friendly technology such as downloadable software for smartphones and electronic prompts to clinicians when guidelines are updated;
Utilization of the intervention options, including access to drugs and devices as delineated by the guidelines;
Outcome assessment, including the extent to which incorporation of a given guideline has benefit to the individual patient from a clinical as well as health resource utilization perspective.
The production and dissemination phases are collectively labor intensive but have been successfully accomplished by members of various health professional organizations over the past 10 years. The most challenging barriers at present thus involve incorporating the most recent guidelines into oncology practice such that patient care is enhanced and measured. A number of reasons create this challenge, including:
The requirement for cancer care clinicians to know of and have facile access to the various guidelines for a given type of cancer management, including oral and gastrointestinal mucositis;
Reluctance on the part of some healthcare providers to change clinical practice due to reliance on the experiential approach they may have utilized for many years;
Limited accessibility to selected drugs and other technology in a given country due to regulatory and/or financial constraints.
Strategies to overcome at least some of these barriers have been incorporated into Web sites such as the National Cancer Institute's PDQ® Web site Oral Complications of Chemotherapy and Head/Neck Radiation (Natl Cancer Inst PDQ®, 2013). This information, presented in companion documents for health professionals as well as patients, was comprehensively updated in March 2011 and included major sections directed to oral mucositis pathobiology and management. In the subsequent twelve months, April 2011–March 2012, this Web site was the 4th of 25 most frequently viewed Supportive and Palliative Care PDQ® Web sites. These types of successes, while important, need to be extended by specifically measuring the clinical and economic impact of guideline utilization in oncology practice, including oral mucositis management.
Key research questions
As described in this review, the field of oral mucositis in oncology patients continues to mature. Reflection upon the past 30+ years reveals an ever-escalating trajectory of advances characterized by (i) basic, translational, and clinical research, (ii) sophisticated modeling of clinical trials, and (iii) increasing collaboration among the researchers and clinicians relative to the production of clinical practice guidelines linked to testing of promising new therapeutics.
There are now new research frontiers to be pursued based on this robust evolution of science and clinical practice. Key research questions for which answers are beginning to emerge are as follows:
What is the role of the oral microbiome relative to causation, progression, and/or healing of oral mucositis?
What are the molecular bases for incidence and severity of pain in relation to degree of oral mucosal injury?
Why do patients with comparable characteristics including age, gender, cancer diagnosis, and treatment regimen collectively exhibit variability of toxicity expression, including oral mucositis?
How can delineation of molecular pathogenesis lead to novel models for prediction of incidence, severity, and response to therapeutics in individual patients, in contrast to a cohort-wide basis?
What biological characteristics contribute to native protection against development of mucositis at selected non-alimentary tract mucosal sites such as vaginal and conjunctival mucosa?
What cross-talk occurs between at-risk mucosal and dermal sites for which common etiologies can be delineated?
How can systems biology technology be optimally applied in order to study the regulatory function of gene clusters in comparison with single-gene expression? In addition, how do these interrelated gene cluster functions translate to studies of symptom burden that are observed in many oncology patients undergoing high-dose cancer therapies?
2.Development of pharmacologics, biologics, and devices to manage oral mucositis:
What are the key molecular targets that, if successfully perturbed, are likely to permit achievement of a high degree of clinical efficacy with least extent of side effects?
What is the most efficient preclinical and clinical development strategy to bring new therapeutics to clinical practice in cost effective fashion?
3.Impact of evidence-based clinical practice guidelines:
What are the most effective ways to ascertain impact of utilization of evidence-based oral mucositis clinical practice guidelines relative to (i) enhanced clinical outcomes and (ii) reduced cost of cancer care?
Given the premier importance of these lines of investigations, these and related key basic and translational research questions were recently addressed at two international conferences:
A first-in-kind Gordon Research Conference in June 2013 titled Mucosal Health & Disease (Gordon Research Conference, 2013). This conference was designed to compare and contrast the science of treatment-induced mucosal disease with naturally occurring mucosal disease such as inflammatory bowel disease and Crohn's disease.
The 6th MASCC/ISOO Mucositis Research Workshop was also conducted in June 2013 (MASCC/ISOO, 2013). This meeting convened clinicians, researchers, and industry professionals to discuss advances in preclinical and clinical mucositis research.
Interdisciplinary conferences such as these are designed to foster new collaborations and innovative strategies to resolving key research questions associated with oral mucositis in patients with cancer.
Recent scientific advances and their clinical application in oncology practice continue to contribute to the maturation of the field of mucosal injury caused by cancer treatments. The historical perspective that oral mucositis was an inevitable toxicity caused solely by basal epithelial injury has been replaced by a sophisticated pathobiological model that is linked to current and developing molecularly targeted interventions and contemporary clinical practice guidelines. New frontiers, including those represented by genomics and proteomics, could lead to prediction of risk of severity of oral mucositis as well as response to therapeutics. These scientific and clinical advances are being further enhanced through coordinated research and health professional efforts at national and international levels in order to fundamentally improve outcomes in patient with cancer while reducing cost of cancer care in the future.
All authors contributed to design and content of the submission.