Cost-effectiveness landscape analysis of treatments addressing xerostomia in patients receiving head and neck radiation therapy

Abstract

Head and neck (H&N) radiation therapy (RT) can induce irreversible damage to the salivary glands thereby causing long-term xerostomia or dry mouth in 68%–85% of the patients. Not only does xerostomia significantly impair patients’ quality-of-life (QOL) but it also has important medical sequelae, incurring high medical and dental costs. In this article, we review various measures to assess xerostomia and evaluate current and emerging solutions to address this condition in H&N cancer patients. These solutions typically seek to accomplish 1 of the 4 objectives: (1) to protect the salivary glands during RT, (2) to stimulate the remaining gland function, (3) to treat the symptoms of xerostomia, or (4) to regenerate the salivary glands. For each treatment, we assess its mechanisms of action, efficacy, safety, clinical utilization, and cost. We conclude that intensity-modulated radiation therapy is both the most widely used prevention approach and the most cost-effective existing solution and we highlight novel and promising techniques on the cost-effectiveness landscape.

In the United States (US), 24 million persons are suffering from xerostomia, or dry mouth, of which, 8 million present with moderate to severe symptoms.^1,2 More than 400 medications are known to be associated with xerostomia as a side effect.^2,3 This is the leading cause of xerostomia which affects in majority the elderly, a population more likely to suffer from chronic diseases necessitating polymedication.³ Other medical etiologies, such as immune syndrome (e.g., Sjögren’s syndrome) and poorly controlled diabetes, can also lead to xerostomia. Xerostomia is the most common complaint of head and neck (H&N) cancer survivors that have received radiation therapy (RT), with a prevalence of 93% during RT and 74%–85% following RT.⁴ Importantly, in those cases, xerostomia cannot be attributed to concomitant chemotherapy (CHT), often used to treat advanced stage cancers, as CHT-induced xerostomia has been shown to be reversible at the end of treatment.⁴ Given 75,000 new H&N cancer patients per year,⁵ 90% of which receive RT⁶ and 85% of which consequently develop xerostomia,^7–9 the incidence of xerostomia as a consequence of H&N RT in the US can be estimated to 30–50,000 new patients each year.

Xerostomia is clinically defined as the subjective complaint of dry mouth and can be related to salivary gland hypofunction, the objective evidence of decrease in salivary secretion (unstimulated whole mouth salivary flow rates <0.1 mL/min or stimulated salivary flow rates <0.7 mL/min).^4,10 However, studies are contradictory as to whether there is an actual relationship between the patient’s subjective perception of dry mouth and the clinician’s objective measure of salivary flow rates.¹¹ Patients might experience xerostomia even without clinical evidence of mouth dryness or hyposalivation, perhaps due to a change in saliva composition. ³ There have been several attempts in the literature to define mild, moderate, and severe xerostomia, according to various subjective and/or objective evaluation criteria (Table I). Despite those attempts, the grading of xerostomia remains nonuniform and there is still no standardized definition. As an example, as recently as 2010, groups have continued to develop and validate new questionnaires to quantify subjective oral dryness (Table I).¹⁴ Other groups have attempted to measure xerostomia based on downstream xerostomia sequelae such as oral pain using the visual analog scale.^15,16 Physicians often use a simple 4-point scale evaluating their patient’s xerostomia, with 0 corresponding to no dryness and 4 corresponding to nonfunctional salivary glands. Under these guidelines, levels 2 and 3 are generally categorized as moderate and severe xerostomia, respectively. However, clinician- reported assessment of xerostomia often drastically underestimates the severity of subjective xerostomia.³² Therefore, to evaluate a patient’s suffering from dry mouth, it is important to assess the following 3 parameters: (1) the subjective feeling of xerostomia symptoms via patient self-administered tests; (2) xerostomia-related quality-of-life (QOL), as evaluated by short questionnaires (Table I); and (3) the clinical evidence of dryness by measuring salivary gland hypofunction using sialometry (objective measures of salivary flow rates).

Table I.

Measures of xerostomia

Type of measure	Name	Year	Performed by	Description	Scale	Ref.
Subjective	Vanderbilt Head and Neck Cancer Survey	2010–2012		28-Item questionnaire, with 5 symptom subscales: “Nutrition,” “Pain,” “Voice,” “Swallow,” and “Mucous/Dry Mouth”	Score from 0–10	12,13
Subjective	Groningen Radiotherapy-Induced Xerostomia questionnaire (GRIX)	2010	Patient	14-Item questionnaire, with 4 subscales: xerostomia during day and night and sticky saliva during day and night	Crohnbach’s α calculated for all subscales is converted to a 0–100 score, higher scores = worse xerostomia	14
Subjective	Visual Analog Scale (VAS)	2002	Patient	Mouth burning and/or pain intensity is evaluated on a 10-cm long VAS	0–10 cm scale, 10 cm being the highest toxicity	15,16
Subjective	Xerostomia-related QOL questionnaire (XQoLQ)	2001	Patient	Five questions relating xerostomia to QOL	Scale from 0–10	Several studies referenced in4
Subjective	Eisbruch’s Xerostomia Questionnaire (XQ), also called University of Michigan XQ (UMXQ)	2001	Patient	8-Item questionnaire evaluating dryness while eating or chewing and while not eating or chewing	0–100 score, higher scores = worse xerostomia	17
Subjective	Xerostomia Inventory (XI)	1999	Patient	11-Item survey	Below 14.5: normal 55: worse toxicity	18,19
Subjective	Patient Benefit Questionnaire (PBQ)	1999	Patient	8-Item questionnaire: difficulty speaking and eating, sleep problems, use of oral comfort aids or fluids, mouth and tongue soreness, and mouth dryness	1–10 Likert scale: 1 = severe negative impact; 10 = no negative impact	18,20
Subjective	Functional assessment of cancer therapy-head and neck (FACT-H&N) questionnaire	1997	Patient	38-Item survey on QOL, 11 of these questions are specific to H&N cancer	QOL score based on the sum of question scores, each rated 0–4 on a Likert scale	18,21
Subjective	Oral Impacts on Daily Performance (OIDP)	1997	Patient	8 Items (eating and enjoying food; speaking and pronouncing; cleaning teeth; sleeping and relaxing; smiling; laughing and showing teeth without embarrassment; maintaining one’s usual emotional state; carrying out one’s major work or social role and enjoying contact with people)	Likert scale for each question, that is summed to a score for each of the 8 categories	22,23
Subjective	Oral Health Impact Profile (OHIP): long form (OHIP49) and short form (OHIP14)	1994, 1997	Patient	49-Item or 14-item (short version) survey, in 7 domains (functional limitation, pain, psychological discomfort, physical disability, psychological disability, social disability, and handicap)	Questions are scored on a 5-point Likert scale and then added to a normalized score	23–25
Subjective	University of Washington Quality of Life questionnaire (UWQoL)	1993–2010	Patient	The questionnaire covers 12 domainsdpain, appearance, activity, recreation, swallowing, chewing, speech, shoulder function, taste, saliva, mood, and anxiety	Questions are scaled from 0 (worst) to 100 (best), and 3 global questions are on a Likert scale (0–5 or 0–6)	26,27
Subjective	Fox’s simple questionnaire (FOX)	1987	Patient	4 Simple questions concerning the patient’s perceptions of oral dryness, oral functions and comfort, and side effects		28
Subjective	Radiation Therapy Oncology Group (RTOG) scoring	1995	Clinician	Evaluating acute xerostomia, based on dryness of mouth and saliva thickness	From 0 (no xerostomia) to 4 (acute salivary gland necrosis)	29
Subjective	RTOG- European Organization for Research and Treatment of Cancer (EORTC) scoring	1995	Clinician	Evaluating late xerostomia, based on mouth dryness evaluation and response to stimulation	From 0 (no xerostomia) to 4 (fibrosis)	29
Subjective	Clinician rating of xerostomia	N/A	Clinician	Based on discussion with patient and/or above questionnaires	0 = no dryness to 3 = nonfunctional salivary	N/A
Objective	Quantitative salivary gland scintigraphy	2000	Clinician	Sequential imaging of the H&N region with a gamma camera after intravenous injection of the radioactive isotope 99mTc-pertechnetate and stimulation of the glands with citric acid	Two measures: maximum tracer uptake within the gland and excretion rate of tracer after stimulation	7,30
Objective	Sialometry: unstimulated whole salivary flow rate (uWSFR)	N/A	Clinician	For 5 min, patient is expectorating periodically into a measuring container	<0.1 mL/min = xerostomia	31
Objective	Sialometry: stimulated salivary flow rate (sSFR)	N/A	Clinician	Patient is chewing paraffin for 5 min, expectorating periodically into a measuring container	<0.7 mL/min = at risk; >1 mL/ min = normal	10
Objective	Clinician rating	N/A	Clinician	Based on uWSFR and sSFR	Grade I (mild), II (moderate), III (severe)	N/A
Objective	Common Terminology Criteria for Adverse Events (CTCAE) v.3.0. scoring	2003	Clinician	Based on a combination of symptomatic evaluation and uWSFR evaluation	Grade 0 (no xerostomia) to 3 (inability to aliment orally, uWSFR < 0.1 mL/min)	29

These various scales and questionnaires can be used to assess a patient’s xerostomia level and they have also been used to evaluate treatment effectiveness in several clinical trials. This list is not exhaustive.

N/A, not available.

Saliva is a complex and versatile bodily fluid that serves a wide range of physiological needs and plays an essential role in oral health. The biochemical composition of saliva includes electrolytes, peptides, proteins, lipids, and antimicrobial substances that coat and protect the oral mucosa from trauma and dehydration, provide an antibacterial, antiviral, and antifungal barrier, maintain a proper pH balance, and prevent demineralization of teeth. Indeed, during early carious lesions, calcium and phosphate ions in saliva can help remineralize the tooth surface. As a consequence, reduction in salivary flow can lead to a broad range of medical sequelae including dental caries, oral infections and sores, difficulties with eating, talking, swallowing, and with function of dental prosthesis as well as altered taste sensation.³³

Xerostomia and salivary gland hypofunction (of all etiologies) have been shown to strongly affects QOL related to daily activities, speech, swallowing, sleeping, and emotional function.^9,34 For example, in a study by Dirix et al., 45% of patients felt that dry mouth invaded every part of their everyday life, 44% of patients felt depressed, and no less than 39% of patients said that dry mouth “diminished their will to live.” In fact, at 6 months post-RT treatment, 80% of patients felt that it was a dire prospect to live with their level of xerostomia for the remainder of their lives.³⁵

The impact of hyposalivation on dental health is of particularly high importance for both dentate and edentulous patients. Dentate xerostomic patients are more likely to have long-term dental and gum disease, extractions, and more invasive dental procedures.³⁴ Therefore, it is expected that higher dental care costs are a direct consequence of xerostomia and that these costs accumulate over time with persistence of hyposalivation and diet change. In the case of Sjögren syndrome-induced xerostomia, Christensen et al.³⁶ demonstrated higher expenses for dental treatment in patients with primary Sjögren syndrome than controls. However, cost of dental care for patients who underwent H&N cancer therapy has not been specifically evaluated, even in studies focusing on the cost of care for H&N.^37,38 Edentulous xerostomia patients may experience extreme discomfort wearing dentures, denture sores, and denture dislodgement (causing social discomfort) because saliva plays a major role in adhesion, cohesion, and surface tension of a denture.³⁹ This clearly impacts QOL and should be further studied. These additional costs of care for dentate xerostomic patients may suggest a need for new guidelines for preventing the consequences of hyposalivation. It may perhaps lead to revisiting the previous recommendations of full mouth extractions for patients undergoing radiotherapy to the oral cavity (instead of the current practice of removing only diseased teeth prior to radiotherapy).

To make a treatment decision for their xerostomia patients, clinicians are contemplating a complex and heterogenous landscape of treatment solutions. Here, we review the current and emerging solutions to address xerostomia in H&N cancer patients and assess their cost-effectiveness. Available treatments for xerostomia typically seek to accomplish 1 of the 3 objectives: (1) to protect the salivary gland during RT, (2) to stimulate the remaining salivary gland function, or (3) to treat the symptoms of xerostomia. Research is also directed toward new methods (4) to regenerate the damaged salivary glands. Within each of these categories, a variety of lifestyle changes, pharmacologic treatments, devices, and surgical procedures are used (Table II).

Table II.

Overview of the competitive landscape: current and emerging solutions to address radiation-induced xerostomia in H&N cancer patients

Type of solution	Type of treatment	Existing solutions	Emerging solutions
Prevention		Intensity-modulated RT Intensity-modulated proton RT Salivary gland transfer Radioprotective drugs (amifostine)	Radioprotective drugs (tempol)
Treatment	Palliation	Use of water Saliva substitutes, gels Hydration pack device
Treatment	Stimulation	Cholinergic muscarinic receptor agonist drugs Chewing-gum and bitter substances Acupuncture	Electrical stimulation devices Gene therapy Growth factors BoNT
Treatment	Regeneration		Stem cell transfer Artificial glands

The solutions as classified according to their goal—prevention or treatment of xerostomia—as well as their stage in the clinical use—existing solutions in the clinics versus emerging solution, still under preclinical or clinical research investigation.

PROTECTION OF THE SALIVARY GLANDS DURING RT

The most attractive way to address xerostomia in H&N cancer patients is to prevent its occurrence in the first place. Several solutions meet this objective: new RT techniques, radioprotective drugs, and surgical procedures (Table III).

Table III.

Current protective treatments

Protecting the salivary glands	IMRT	Amifostine	Salivary gland transfer
% Patients still   experiencing   xerostomia post-RT	25%–40%8,40	51%41	~20%42
Treatment planning time	15–26 h43	None	None
Treatment delivery time	15–30 min44 per session, in 35 sessions over 7 weeks	None	45 min45
Cost calculation	2006 MP for whole breast IMRT = $29,79046 2010 MP for prostate cancer IMRT = $42K47 2011 MP for: IMRT planning, (CPT code 77301),   $2088.19 þ IMRT delivery (CPT code 77418),   $519.84 per treatment session × 35   sessions = $20,282.5948	Calculated assuming the patients gets 8 mg of   odansetron (anti-nausea drug) before each of 35   subcutaneous injections of 500 mg amifostine, before   each RT session, over a 7-week period. 2011 MP for: odansetron (HCPCS code Q1079),   $0.676 × 35 sessions + amifostine (HCPCS code   J0207), $322.019 × 35 sessions = $11,30049	2010 MP for “unlisted procedure, salivary glands or   ducts” (CPT code: 42699), $1335 (RVU = 0)*; 2010   MP for “excision of submandibular glands” (CPT   code: 42440), $1725 (RVU = 7.13)*
Cost estimate per   treatment   course	$20-$100K	$10-$30K	$10-$20K
Invasiveness	None	None	High
Contraindications	None	Allergies to aminothiol compounds	Oral cavity malignancies requiring RT to submental   area + any H&N cancer not requiring surgical   resection
Side effects	None	Nausea, vomiting, hypotension	None

Protecting the salivary glands	BoNT (emerging treatment option)	IGF-1 growth factor (emerging treatment option)
% Patients still experiencing   xerostomia post-RT	N/A. Preclinical stage	N/A. Preclinical stage
Treatment Planning time	None	None
Treatment delivery time	1 min	1 min
Cost calculation	Less than 20 U of BoNT per patient will be necessary and up to 100 U per bottle may be   charged if 1 bottle is used for only 1 patient, amounting the cost of the drug to   $125-$600 (BoNT type A, HCPCS code: J0585, $5.48/U).49,50 Treatment delivered   in a single injection, estimated to $200, based on CPT codes for a similar procedure   (CPT code 64613 for injection of BoNT to the neck to treat muscular pain, is $164.11   in non-facility setting),48,50	Based on an FDA-approved daily dose of 0.24 mg/kg,   adapted to a 70 kg adult and an average 2006   wholesale price of $562.50 per vial of 40 mg   Increlex,51 the annual cost of an IGF-1 treatment for   xerostomia can be expected to be in the $100K range
Cost estimate per treatment course	$325-$800	$100K
Invasiveness	None	None
Contraindications	None	Unknown
Side effects	None	N/A

The current treatments aimed at protecting the salivary glands are summarized. For each treatment, details are provided about the following parameters: xerostomia rate after RT, treatment planning duration, treatment delivery duration, cost estimate per treatment course, with detailed cost calculation, invasiveness, contraindications, and side-effects.

MP, medicare payment; RVU, relative value unit; N/A, not available.

^*2010 Otolaryngology fee schedule.

Newer RT techniques are aimed at delivering a spatially selective therapy, which minimizes dosage to normal tissues when irradiating cancerous lesions. Introduced around 2000, intensity-modulated radiation therapy (IMRT) achieves this goal by creating a highly sculpted radiation dose distribution using multiple beams, each with its own spatially varying intensity profile.⁷ The clinical approach is as follows: wearing a custom-fit thermoplastic immobilization mask, the patient undergoes CT imaging, which will be used by the radiation oncologist to (1) delineate tumor volumes and margins and (2) compute beam intensities. Subsequently, the treatment is delivered in approximately 35 daily fractions over 7 weeks. IMRT can better spare the parotid gland adjacent to crucial lymphatics that need to receive a high dose of radiation.⁷ IMRT results in a mean parotid radiation dose of 30 Gy,⁵² a dose close to the threshold of 26 Gy suggested for severe parotid injury.⁷ Due to the proximity of submandibular salivary glands to the primary tumor and/or crucial lymphatics in the oral cavity (lymph node group 1B or submandibular triangle), as well as due to primary tumor size, IMRT often might not be able to effectively protect these glands, which are responsible for unstimulated saliva production. Consequently, in many cases, IMRT may not be able to protect against xerostomia at rest, especially at night. In selected patients harboring more lateralized primary tumors, recent studies have demonstrated that it is possible to spare the contralateral submandibular gland using IMRT.^7,53,54 However, this practice is limited to early stage cancers that do not require radiation to bilateral necks. Most advanced cancers (stages 3–4) require bilateral necks irradiation and the addition of CHT to the radiation, potentiating the effects of radiotherapy to eliminate cancer cells. As a side effect, this synergistic effect might also cause more damage to salivary tissue resulting in xerostomia. A study is ongoing combining the surgical submandibular gland transfer technique (see later description of the Seikaly-Jha procedure for submandibular salivary gland transfer) with tomotherapy-based IMRT,⁵⁵ a very promising solution, which might provide a better sparing of the submandibular glands, but remains invasive, time-consuming, and very costly. Newer rotational IMRT techniques, such as volumetric modulated arc therapy and helical tomotherapy, can improve delivered radiation dose, reduce radiation delivery time, and carry potential for a better conformation to the salivary glands: preliminary results in terms of salivary function seem encouraging,^56,57 but there is still a debate as to whether the dosimetric improvements translate into clinical outcome, as compared with standard step-and-shoot IMRT.⁵⁸ As a consequence, IMRT certainly reduces xerostomia as per patient survey—from 80% down to 25%–40% incidence, according to 3 recent prospective randomized studies conducted in the UK and in Hong Kong.^7,59–61 This has been confirmed in a systematic review by the Oral Care Study Group of the Multinational Association of Supportive Care in Cancer and International Society of Oral Oncology (MASCC/ISOO) demonstrating that parotid sparing IMRT has the potential to reduce the prevalence and severity of xerostomia and improve xerostomia-related QOL after therapy.⁴ On the practical side, IMRT is very labor and time intensive and requires a skilled radiation oncologist, radiation physicist, and dosimetrist. IMRT planning is lasting 5–16 h longer than typical RT planning (~10 h) and daily IMRT treatment delivery lasts approximately 20–30 min. IMRT is currently the recommended standard treatment for H&N cancer⁸ Indeed, parotid-sparing IMRT was recommended by the MASCC/ISOO for the prevention of xerostomia and salivary gland hypofunction.⁹ It has been shown that saliva secretion has the potential of increasing following IMRT,⁹ suggesting that sparing of salivary glands by IMRT could make them more amenable to treatments stimulating their function (see later). IMRT is available at most RT centers in the US, where it is increasingly used. As an example, almost 90% of H&N cancer patients are treated with IMRT at our institution’s Hospital and Clinics. However, it is only available to 10% of the global population.⁴⁰ We estimated the cost of H&N IMRT treatment to $20K-$40K (Table III). However, we acknowledge that this is a simplified cost addressing the whole IMRT treatment including planning time for multiple vital structure sparing and tumor coverage, as it is very challenging to estimate cost applicable only to salivary gland sparing.

Proton RT can potentially achieve a higher sparing of salivary glands than IMRT due to a more localized radiation dose delivery. Although there are dosimetric studies examining protons for salivary gland sparing,^62,63 there is no clinical trials for the use of Proton RT in H&N cancer management.

In 1995, amifostine (Ethyol; MedImmune Pharma, Nijmegen, The Netherlands), a radioprotective drug, was approved by the Food and Drug Administration (FDA) for the prevention of xerostomia in patients undergoing H&N RT. This pharmacologic treatment mainly protects the parotid glands from radiation-induced apoptotic death by acting as an oxygen-radical scavenger. The treatment is administered by subcutaneous injection (500 mg) before each RT session for 6 weeks.⁶⁴ Amifostine significantly decreases xerostomia by 22%–27% up to 12 months post-RT.^36,119 However, amifostine has many detrimental side effects, the major one being nausea/vomiting,⁴¹ mandating the concomitant use of a nausea-preventing drug. Still, 21% of patients drop out of amifostine treatment regimens due to the side effects.⁶⁵ We estimate the amifostine treatment cost to $10K-$30K (Table III). Amifostine does not specifically target salivary glands, so there is a theoretical concern that amifostine could protect the tumor from radiation through the same mechanism by which it protects the salivary glands; however, there is no evidence in favor of this hypothesis.⁹ The MASCC/ISOO Study Group could not recommend any guidelines for the use of amifostine to prevent xerostomia during RT, due to lack of consensus on the interpretation of existing evidence.⁹ Many oncologists also avoid prescribing amifostine because of the lack of agreement that the benefits of salivary glands protection are outweighing the side effects.

Botulinum toxin (BoNT) is a promising emerging pharmacologic treatment for radiation-induced xerostomia. A preclinical rat model has recently demonstrated that intraglandular injection of BoNT before RT reduces glandular injury.⁶⁶ The mechanism of action of this treatment is unclear and two hypotheses have been proposed: (1) a temporary glandular involution leading to reduction of saliva production during RT, avoiding concentration of radiation where inorganic solutes from saliva are located or (2) an action on the nitric oxide pathway. Furthermore, BoNT has been shown to increase tumor response to RT in an experimental model, suggesting that this treatment would not have the risk of tumor protection.⁶⁷ Because BoNT is already an FDA-approved drug for many applications, clinicians would more likely be inclined to prescribe this potential cytoprotective treatment before the start of radiotherapy, and the regulatory hurdles are correspondingly lower. We can estimate the total treatment cost to $325-$800 (Table III).

Other emerging preventive treatments include systemic administration of growth factors like insulin growth factor 1 (IGF-1) or keratinocyte growth factor, both of which have been shown to preserve gland functions in preclinical mouse studies.^68,69 These growth factors are believed to protect the salivary glands by two possible mechanisms: (1) improving the survival and proliferation of salivary acinar cells and stem cells and (2) suppressing apoptosis of those cells after radiation.⁸ FDA approval has not been granted for IGF-1 injections in humans for several major applications (diabetes and amyotrophic lateral sclerosis). The only approval given to date for this therapy is for severe IGF-1 deficiency inducing dwarfism (Increlex (mecasermin); Ipsen, Paris, France),⁷⁰ so it seems unlikely to receive approval for xerostomia treatment. In addition, radiation oncologists are cautious in using growth factors as a preventive agent for xerostomia because of a possible protumor effect. Treatment cost is estimated in the $100K range (Table III).

Another emerging radioprotective drug is tempol (4-hydroxy-2,2,6,6-tetramethylpiperidine 1-oxyl; Sigma-Aldrich, St. Louis, MO, USA), which has been shown, after systemic or topical application, to protect salivary glands but not tumor tissue from radiation damage in a mouse model⁷¹ tempol is a stable nitroxide that has several possible mechanisms of action including (1) mimicking superoxide dismutase activity, (2) oxidizing transition metals, and (3) scavenging free radicals.⁸ Tempol is a promising drug: it is currently tested in its gel formulation, also called MTS-01, in Phase-II clinical trials for the treatment of alopecia induced by brain RT and for the treatment of dermatitis induced by radiation and CHT for anal cancer.⁷² However, a clinical trial on the use of tempol to treat xerostomia has not yet been initiated.

An approach to surgical preventive care for xerostomia is the Seikaly-Jha procedure for submandibular salivary gland transfer. In this procedure, 1 submandibular salivary gland is surgically transferred into the submental space on the contralateral side of the primary tumor, which is typically shielded during RT.⁴² The surgical procedure is safe, does not require any microvasculature expertise, and does not involve a dedicated surgery because it can be performed during surgical resection of the primary tumor. Performing the Seikaly-Jha procedure adds approximately 45 min to the 2–10 h total surgical time,⁷³ and the level of sparing obtained by this procedure has been suggested by the MASCC/ISOO study group to possibly be of clinical significance. ⁹ It has been reported to prevent xerostomia in 83% of patients in the long term (2 years).⁴² However, this invasive procedure is typically only applicable to patients with specific cancers requiring surgical resection as part of their H&N treatment (e.g., oropharyngeal carcinoma). It is also contraindicated in oral cavity malignancies because the submandibular nodal basin and the submental space are often irradiated in that case. This procedure was developed in Canada and has been clinically tested in Canada⁷⁴ and China.⁷⁵ It is currently under clinical trial in the US under the auspices of the Radiation Therapy Oncology Group.⁷⁶ Information on cost is unavailable but we estimated it to $2-$20K (Table III).

STIMULATION OF THE SALIVARY GLANDS

In many cases, prevention of xerostomia cannot be achieved efficiently (e.g., because the salivary glands are located close to the primary tumor or lymph nodes suspected of tumor involvement) and the patient is left with hypofunctioning salivary glands. Data pooled from a large number of clinical studies has shown that, during or following RT, stimulated saliva secretion is consistently higher than unstimulated secretion, suggesting that stimulation of the salivary glands might be an effective strategy to treat xerostomia.⁴ Several pharmaceutical, alternative medicine, or devices treatments aim at stimulating the glands to “squeeze out” more saliva in these instances (Table IV).

Table IV.

Current stimulation and palliation treatments

Stimulation and palliation	Pilocarpine or Cevimeline	Acupuncture	Saliva substitutes	Hydration pack (Xeros)	Electrical stimulation (GenNarino)
% Patients experiencing partial reduction in xerostomia	80%77	55%78	40%8	Not documented	N/A
Treatment duration	Lifelong	2 × 20 min/week for 4 weeks18	Lifelong	Lifelong	Lifelong
Cost calculation	3–4 of 5 mg Salagen pills per day, costing $3.95-$5.27 per day. Or 4 of 30 mg Evoxac capsules per day, costing $3.93/day79	Average cost of an acupuncture session, $50-$7080 × 8 sessions	40–150 mL of saliva substitute per day81 × $0.2/mL approximate US retail price for Biotene Oral Balance Liquid	Device retail price = $79982	Device retail price = $575†
Cost estimate per treatment*	$5/day	$400-$600	$0.5-$2/day	$79982	$575†
Invasiveness	None	Minimally	None	None	None
Contraindications	Asthma, chronic obstructive pulmonary disease, cardiovascular disease, and glaucoma7	N/R	N/R	N/R	N/R
Side effects	Sweating, hot flashes7	N/R	Sticky mouth, mucosal damage	N/D	N/R

The current treatments aimed at stimulating the salivary glands or treating xerostomia symptoms are summarized. For each treatment, details are provided about the following parameters: xerostomia rate after treatment, treatment duration, cost estimate per treatment course, with detailed cost calculation, invasiveness, contraindications, and side effects.

N/A, not available, N/D, not documented, N/R, none reported.

^*Cost estimates are for the entire treatment duration, except for lifelong treatments, where the cost estimate is given on a daily basis.

^†A. Wolff, President of Saliwell, Ltd, written communication.

There is a long list of pharmacologic sialogogues (drugs that increase saliva flow) but only 2 are approved for xerostomia treatment: pilocarpine hydrochloride (Salagen; MGI Pharma, Bloomington, MN, USA) and cevimeline (Evoxac; Daiichi-Sankyo, Tokyo, Japan).⁸ These cholinergic muscarinic receptor agonists mimic saliva-inducing nervous signals. In the case of pilocarpine, the mechanisms of action are not fully understood and might be attributed to stimulatory action on minor salivary glands.⁹ The results of several studies with pilocarpine have been inconsistent,^83,84 and the beneficial effect of both drugs on xerostomia is debated. Although salivary flow increase has been shown,⁸⁵ there may be no difference in the subjective perception of xerostomia.⁸ Among the 42%–51% of patients responding to pilocarpine, it was shown that the response can be delayed up to 12 weeks.⁸⁶ Furthermore, the positive effects disappear as soon as the patient stops the treatment, so these drugs would need to be taken for a patient’s entire lifetime. Both drugs have many contraindications, such as asthma, and moderate adverse effects including intense sweating and hot flashes (Table IV). The treatment is administered orally as a dose of 5 mg 3 times daily and costs $4-$5/day (Table IV). The MASCC/ISOO group cannot recommend the use of pilocarpine during RT due to “equivocal results of the various randomized clinical trials” but, however, recommends its use following RT to improve xerostomia.⁹ Oncologists routinely prescribe pilocarpine to patients complaining of xerostomia but many patients (6%–15%)⁸⁶ drop out of the pilocarpine treatments because of the low benefit-to-cost ratio, both in terms of side effects and treatment cost.

Acupuncture, an “alternative medicine” treatment believed to work through neuronal activation,⁸⁷ has been shown to stimulate residual salivary gland function. Pilot studies have demonstrated a sustained effect on saliva secretion for up to 3 years and improvements in subjective symptoms (55% decrease in xerostomia in IMRT patients).⁷⁸ A related proprietary technique called Codetron (Ehm Rehabilitation Technologies, Inc., Ontario, Canada) based on transcutaneous electrical nerve stimulation has showed 6-month improvements in whole saliva production.⁸⁸ However, acupuncture and related techniques might not be considered a clinical treatment option by clinicians and their scientific value may be subject to debate. Nevertheless, the use of acupuncture to stimulate salivary secretion and alleviate xerostomia was suggested by the MASCC/ISOO study group.⁹ The cost is estimated at $400-$600 per treatment course based on current practice rates in the US (Table IV).

Electrical salivary stimulators are emerging devices that can be fixed or placed inside the mouth and apply electrical currents to the nervous extensions in the oral mucosa that innervate the salivary glands and stimulate salivation.^89,90 Among them, the Salitron device (Biosonics, Fort Washington, PA, USA) is an intra-oral handheld stimulation probe linked to a current generator that was FDA approved in 1988. This cumbersome apparatus caused a sustained increase of salivary flow rate, as well as subjective improvement in xerostomia symptoms.⁹¹ Because this device was limited by its large size, high price, and lack of user-friendliness, it was replaced by miniaturized devices: GenNarino (Saliwell Ltd, Saarbruecken, Germany), a remotecontrolled removable intra-oral splint appliance⁹² and Saliwell Crown (Saliwell Ltd), an osteointegrated dental implant.⁹³ Short-term (10 min) higher mouth moistening and reduced xerostomia have been demonstrated in both cases.⁹² Clinical trials are ongoing to evaluate the long-term effect of these devices on xerostomia.⁹⁴ Saliwell GenNarino has been granted approval for marketing in Europe (CE mark) and is sold at a price of $575 per device (by A. Wolff, President of Saliwell, Ltd, written communication, February 2011).

Small studies have evaluated the use of gustatory and masticatory stimulants (gums, acidic candy, and salivastimulating lozenges) for stimulating salivary gland function in xerostomia patients following RT, but the results are not consistent: although chewing gum can increase saliva production for patients with remaining salivary function, there is no evidence that gum is more or less effective than salivary substitutes (see later), which do not seem to be more efficient than placebo.³ Therefore, no consensus can be extracted from the results of these studies.⁹ Furthermore, the use of gum induces stickiness of the mouth, and the use of acidic candy can cause erosion of the tooth enamel.³

REDUCTION OF XEROSTOMIA SYMPTOMS

In most cases of moderate to severe xerostomia following RT, the remaining nonfunctioning or hypofunctioning salivary glands cannot be stimulated efficiently to alleviate the subjective feeling of dry mouth. In those instances, supportive and palliative treatments aimed at moistening the mouth are the only options (Table IV).

Clinicians recommend their xerostomia patients adopt a change of lifestyle to treat their symptoms. This includes the frequent use of water, sucking of ice chips, antibacterial saliva substitutes, and moistening agents to palliate mouth dryness, together with avoidance of irritants such as spicy foods, alcohol, caffeine, or smoking.³ Saliva substitutes are solutions that mimic the essential properties of normal saliva, including its viscosity, lubrication, wetting properties, and antimicrobial effects.⁹⁵ The majority of those solutions, based on mucin, carboxymethylcellulose, hydroxyethylcellulose, or xanthan gum,⁹⁵ seem to relieve xerostomia for approximately 40% of patients,⁹⁶ but the relief is only temporary.⁶⁷ Oral gel formulations, which harbor a thicker texture that can line up oral mucosa and enamel, provide a longer lasting moisture sensation and are recommended for night use.⁸ Under these palliative treatments, xerostomia patients still have to wake up many times at night to reapply the treatment, drink water, or go to the bathroom due to the resulting polyuria. This short lasting effect is possibly the reason why the MASCC/ISOO panel recommended the use of these lubricants and saliva substitutes for short-term xerostomia improvement following RT.⁹ Conversely, the Cochrane review on the subject states that there is insufficient evidence that saliva substitutes are better or worse than placebo in reducing xerostomia symptoms and does not provide recommendations on this matter.³ Large volumes of saliva substitute need to be applied every day: an average of 40 mL for mucin- and 150 mL for carboxymethylcellulose-based saliva substitutes, as reported by Vissink et al.⁸¹ Except for the lifelong use of water, the other solutions are expensive, in the range of several dollars per day (Table IV). Moreover, a strong compliance is needed for treatments that require frequent applications.

Reservoir-based devices such as a hydration device called the Xeros hydration pack (Lorin Technologies Corporation, Swansboro, NC, USA) are trying to address this compliance issue. The device is an automated pump system for delivering water or saliva substitute to the mouth from a reservoir of liquid carried in a fanny pack via tubes going to the patient’s mouth. This device, previously commercialized for hydration during hiking, received FDA approval in April 2011 and is currently commercialized for xerostomia treatment at a price of $799 per device.⁸² Other oral reservoir devices (mouthguards or dentures) have been clinically tested in pilot trials,^97–99 but there is insufficient evidence at present to recommend their use,³ and information on commercialization or cost of these devices is unavailable.

REGENERATION OF THE SALIVARY GLANDS

As an attempt to reverse the damage caused by RT to the salivary glands, several research strategies are emerging, aimed at regenerating salivary glands either via gene therapy or by transplantation of salivary gland stem cells. Devices containing saliva-secreting cells are also envisioned.

Preclinical studies in rats and miniature pigs have demonstrated that gene therapy using the gene coding for a transmembrane water channel protein, human aquaporin-1 (hAQP1), helps recover salivary function by up to 80% of pre-irradiation baseline saliva production.^100–102 The mechanism of action is believed to be an increase in water secretion by duct cells in the salivary glands. However, this adenoviral (adeno-associated virus, AAV) treatment has been shown to induce an inflammatory response in the targeted salivary glands in a preclinical model,^103,104 an effect that might be reversible.¹⁰⁵ Recent studies showed that AAV transfer of the human keratinocyte growth factor gene reduced postirradiation xerostomia in a mouse model,¹⁰⁶ whereas AAV transfer of tousled-like kinase 1B, a prosurvival gene, helped prevent radiationinduced damage to the salivary glands in a rat model.¹⁰⁷ A new method, based on ultrasound-assisted gene transfer has recently demonstrated that those genes could be transferred to the salivary glands without the use of a viral vector.¹⁰⁸ The only clinical trial for gene transfer to the salivary glands is a phase I clinical trial that will determine the safety and effectiveness of an adenoviral (AAV) therapy encoding hAQP1 in humans AdhAQP1.^101,109,110 However, obtaining FDA-approval for such a therapy might be very difficult, especially because the outcomes of previous gene therapy trials have revealed a risk of fatal outcome.¹¹¹ The cost of gene therapy is extremely high, in the order of $100K/year.¹¹²

Recently, the potential of specific stem cells identified by their expression of several progenitor markers to regenerate salivary glands was reported using a radiation-damage preclinical model.^113,114 Salivary gland stem cells research holds tremendous promise for the future of xerostomia research.¹¹⁵ However, results are still preliminary and stem cells are not expected to be used in the clinics in the near future.

Several patents have been filed regarding artificial salivary glands consisting of biomaterial scaffolds containing cells capable of secreting saliva.^116–118 Like the stem cell approach, the research related to this treatment option is still very immature and not likely to be applicable in the clinics in the near future.

DISCUSSION AND CONCLUSION

Although partial recovery may happen during the first year after H&N RT, the majority of H&N cancer patients experience long-term xerostomia, significantly impairing their QOL and potentially incurring extensive dental care costs. Here we presented an overview of the different treatment solutions for radiation-induced xerostomia, summarized as a cost-effectiveness gap analysis (Figure 1). These solutions have very heterogenous utilization profiles, with half of them still in preclinical or clinical trials and the other half being currently used or recommended in clinical practice: IMRT, proton therapy, various oral lubricants, pilocarpine, amifostine, and hydration and electrical stimulation devices. The degree of adoption of each of these treatments can vary a lot, depending on their cost, availability, and risk-to-benefit ratio. Prescribing pharmacologic treatments has demonstrated only moderate results, counterbalanced by possible QOL-impairing side effects (e.g., pilocarpine). We conclude that the most widely used strategy in the clinics is a combination of xerostomia prevention by IMRT followed by post-RT treatment of xerostomia symptoms by frequent hydration and saliva substitutes. Due to their antibacterial properties, these saliva substitutes have the potential to reduce the high dental care costs incurred by xerostomia patients. However, due to relatively short lasting moisturization, these solutions possibly have a poor efficacy at treating xerostomia at resting state and nighttime.

Fig. 1.

Treatment gap analysis. Synthesis of the solutions for addressing xerostomia in H&N cancer patients undergoing RT. The solutions are summarized in a graph showing the effectiveness of the solution at treating or preventing xerostomia versus the cost per treatment. The solutions are categorized according to the following criteria: treatment versus prevention and preclinical versus clinical stage. Relative clinical availability is also estimated and represented as bubble size. Highlighted by the dashed circle is the treatment gap, which corresponds to the area of greatest innovation opportunity.

The solutions described here occupy very different locations on the cost-effectiveness landscape (Figure 1). In our analysis, we observed a gap in treating xerostomia in H&N cancer patients with both a high effectiveness and a relatively low cost. BoNT and acupuncture treatments appear as the most cost-effective existing strategies and seem very promising because both techniques are already widely available for different applications. However, these are emerging solutions still being tested in preclinical and clinical research respectively: their efficacy at treating xerostomia need to be further studied and confirmed. IMRT appear to be the most cost-effective existing solution that is already widely available and in use in the clinics. Surgical gland transfer harbors a similar cost-effectiveness ratio and could be easily implemented in the clinics (during tumor resection surgery, when applicable) but remains very invasive and costly. Combining those two techniques seems very promising but will likely incur very high costs.

It is important to note that the cost of care for patients suffering from hyposalivation also includes extensive dental care costs, which could potentially be reduced by the solutions mentioned above. Furthermore, those costs could be further reduced if H&N cancer patients are treated in multidisciplinary settings where cancer care and dental care are integrated.

On the basis of our analysis, the ideal solution addressing resting-state xerostomia in H&N cancer patients is still missing, which would be able to reduce the perception of mouth dryness at resting state, achieve the protective function of saliva, be noninvasive, and cost-effective, at least with respect to current pharmaceuticals (in the order of $1-$2/day). In our opinion, the current clinical landscape is lacking such a medical solution and there is room for innovators to design novel solutions to meet the need of RT-induced xerostomia in H&N cancer patients. It is worth noticing that such a solution would have the potential to serve the needs and improve QOL of H&N cancer patients suffering from radiation-induced xerostomia, as well as a broader population of patients suffering from xerostomia of various etiologies, such as Sjögren’s syndrome and polypharmacy.

Statement of Clinical Relevance.

Xerostomia is the major complaint of patients receiving H&N RT. This manuscript is a practical and comprehensive review of solutions for prevention and treatment of radiation-induced xerostomia with emphasis on cost-effectiveness to help guide clinical treatment decisions.