27

Salivary gland secretion is controlled via neurotransmitters under the influence of the autonomic nervous system, although various hormones may also modulate salivary composition. In general, parasympathetic stimulation increases fluid secretion as a result of the activation of M3 muscarinic receptors on acinar cells; sympathetic stimulation via alpha-1-adrenergic receptors also produces more saliva, though much less than that occurs after muscarinic stimulation; and stimulation via beta-adrenergic receptors stimulates salivary protein release from acinar cells via fusion of zymogen granules. Neuropeptides (substance P and vasoactive intestinal peptide) and autacoids (histamine and bradykinin) may also influence salivary secretion. Water-specific channels or aquaporins facilitate water movement across acinar cell plasma membranes and provide the only source of fluid secretion in salivary glands.

The most obvious function of saliva water, mucins, and proline-rich glycoproteins is to lubricate food and mucosa and help taste perception and swallowing, but other functions include the following (Table 27.1):

• Digestive: Salivary amylase has a very minor role in humans in the conversion of starch to maltose. Salivary lipase may assist fat digestion.

• Excretory: Some drugs, such as alcohol, are excreted in saliva. Secreted cancer chemotherapy agents may lead to mucosal toxicity, and some may result in altered taste.

• Maintenance of tooth integrity: The buffering capacity and supersaturated calcium and phosphate are important in maintaining tooth integrity.

• Hormonal: EGF (urogastrone), a polypeptide from submandibular glands (SMGs), may play a role in wound healing. Homeostatic proteases, such as kallikrein, renin, and tonin, may control local vascularity and water/electrolyte transport.

• Protective: The lubricative and mechanical washing effects of saliva as well as nonspecific and specific immune-protective mechanisms protect the host. Saliva is inhibitory to various microbial agents, including human immunodeficiency virus (HIV).

These mechanisms include the following:

• Mucins that aid lubrication, aggregate bacteria, are antiviral, and limit mucosal permeability to various toxins.

• Inhibitors of proteolytic enzymes, such as cysteine-containing phosphoproteins and antileukoproteases that are, with mucins, protective against proteolytic enzymes from bacteria and leukocytes.

• Bacterial aggregators that can aggregate bacteria and prevent their attachment to oral surfaces, such as mucins, some glycoproteins, and lysozyme.

• Direct nonimmune antimicrobial mechanisms, such as defensins,

Lysozyme interacts with anions such as thiocyanate to lyse gram-positive bacteria.

Defensins and histidine-rich peptides in parotid saliva also suppress oral bacteria and fungi.

Lactoferrin chelates iron and deprives bacteria of an essential factor.

Peroxidase with thiocyanate and hydrogen peroxidase acts against some gram-positive and gram-negative bacteria and yeasts.

Amylase may, for example, protect against Neisseria gonorrhoeae.

• Immune protection—principally via secretory immunoglobulin A antibodies.

Table 27.1 Functions of Saliva

Treatment of head and neck cancer (HNC) and bone marrow (hematopoietic stem cell) transplants may especially be associated with xerostomia1,2; in patients with cancer, xerostomia has independent negative influences on the quality of life (QoL).^3,4 This also applies to children with malignant disease⁵ and in patients who receive chemotherapy for solid tumors.⁶ Furthermore, patients with advanced cancer frequently have xerostomia and dry mouth, which is commonly associated with oral discomfort and dysphagia.⁷ This chapter summarizes the area and highlight recent advances and future directions: several recent reviews have covered this field comprehensively.⁸

The other main causes of hyposalivation are drugs (those with anticholinergic or sympathomimetic activity), irradiation of the major salivary glands, Sjögren syndrome, diabetes, HIV infection, sarcoidosis, and dehydration (Table 27.2).

Cancer Therapy Effects on Salivation

While the direct effects of some cancer therapies on salivary function frequently cause hyposalivation, 18 to 19% of both hospitalized patients with cancer and patients without cancer may suffer from dry mouth,⁹ suggesting a strong role of medications and anxiety and/or depression in hospitalized patients.

Effects of Cancer Therapy

In addition to saliva production, the quality of saliva is frequently affected. Increased viscosity produces symptoms during cancer therapy, which may become a chronic complaint. The mucous acini of salivary glands have a reduced sensitivity to toxicity, and as serous secretions decline, they may retain function for some time. The production of excessively thickened secretions affects the flow of the secretion and may lead to nausea and vomiting, particularly in patients receiving chemotherapy that places them at increased risk for nausea.

Irradiation of the major salivary glands is common in the treatment of cancers of the head and neck (H&N), thyroid, and lymphomas. In HNC, when the salivary glands and bilateral neck irradiation is required, hyposalivation occurs. Hematopoietic stem cell transplantation also involves damage to salivary function,10 particularly when total body irradiation is part of the conditioning protocol and if graft-versus-host disease develops and involves the salivary glands.¹¹

Table 27.2 Causes of Xerostomia

Radiotherapy (RT) causes acinar cell apoptosis, leading to a change in saliva quantity and quality in approximately 1 to 2 weeks of beginning RT at a cumulative dose of approximately 10 Gy; salivary function falls as the RT dose increases, and at a total dose of more than 50 Gy, virtually complete hyposalivation can follow when all glands are in the RT field.12,13 Salivary flow rates fall dramatically during the first 2 weeks of RT, and both the parotid and submandibular/sublingual glands can be similarly affected.¹⁴

Patients with HNC treated with RT either alone or in combination with chemotherapy or surgery report xerostomia as one of the most frequent complaints, and this has a significant effect on the more general dimensions of health-related QoL.¹⁵ In HNC, xerostomia increases from 19% pretreatment to 62.6% during RT and 53.2% after RT.¹⁶ In patients treated for nasopharyngeal carcinoma, xerostomia persisted at the last follow-up (24 months).¹⁷

Salivary function recovery may occur within 1 year after RT¹⁸; however, little improvement can be expected after 1 year following cancer treatment, and dry mouth is the most common chronic complaint of patients after HNC therapy that includes radiation therapy.

Diffusion-weighted magnetic resonance imaging allows noninvasive evaluation of functional changes in the major salivary glands after RT and is a promising tool for investigating RT-induced xerostomia.¹⁹

Effects on Quality of Life

The assessment of QoL in HNC includes the Head and Neck Symptom Scale of the University of Washington Quality of Life (UW-QoL) questionnaire, which includes items related to saliva amount and consistency (Table 27.3).²⁰

The University of Michigan Xerostomia-Related Quality of Life Scale²¹ is a 15-question survey, with each response based on a 5-point severity response. Others have used broad QoL questionnaires with modifications or additional sections. Dry mouth has a significant effect on overall QoL.²² Two hundred and eighty eight patients with all stages of HNC were assessed by using the European Organization for Research and Treatment Core QoL Questionnaire (EORTC QLQ-C30) and Radiation Therapy Oncology Group (RTOG) toxicity criteria up to 24 months after cancer therapy.

Table 27.3 Scoring Salivary Function

Xerostomia was found to have a significant effect on overall QoL (p < 0.001) and the effect on QoL increased over time.22 One hundred and forty nine patients with stage III or IV HNC were assessed pretreatment and at 1 year after treatment by using EORTC QLQ-C30 and EORTC H&N35. The primary complaints were dysphagia, dry mouth, and thick saliva (p < 0.05), and patients with oral cancer had limited improvement at the last follow-up.²³

A study of 65 patients with HNC who had completed RT more than 6 months earlier showed the most common chronic symptoms—dry mouth (92%), change in taste (75%), difficulty in swallowing (63%), moderate to severe difficulty in chewing (43%), and sore mouth when eating (40%).²⁴ A prospective study of 357 patients with HNC identified chronic oral symptoms—dental problems, trismus, xerostomia, and sticky saliva that persisted or increased over time after 1 year and persisted for 5 years.²⁵

EORTC QLQ-C30 has been used for the assessment of QoL, and specific addenda addressing oral symptoms and effect on QoL have been used in several studies.²⁶

Other tools have been assessed, including a visual analogue scale (VAS) for subjective assessment of saliva and a reliability of seven of eight VAS responses, which were found to predict changes in saliva flow induced by xerostomic medications.²⁷

A prospective, multicenter study of QoL conducted in 122 patients with oral cancer (62% man, mean age 61 years) with patient reported outcomes completed pretreatment, and at 1 and 5 years after treatment,²⁸ it found oral complaints to include dry mouth, sticky saliva, speech changes, dental problems, and sleep disturbance. Symptom burden remained significant at 5 years after treatment in patients who were treated with RT, and these complaints were associated with decreased QoL (p < 0.01). In another study, 107 patients completed QoL surveys before and 6, 12, 24, and 36 months after HNC treatment.²⁹ Most short-term morbidity resolved in 1 year of cancer treatment. At the end of follow-up, physical functioning, taste/smell, dry mouth, and sticky saliva were significantly worse compared with those at baseline.

Several studies assessing chronic symptoms more than 6 months to 5 years following Tradiotherapy for HNC have shown that dry mouth, thick saliva, and dysphagia are the most common and troubling persisting complaints.^28,30,31 Most short-term morbidity resolved in 1 year of cancer treatment. At the end of follow-up physical functioning, taste/smell, dry mouth and sticky saliva were significantly worse compared with baseline.

Minimizing Radiation-Induced Xerostomia

Several strategies are available to minimize radiation damage to salivary glands (Table 27.4).

Table 27.4 Strategies to Minimize Radiation-Induced Xerostomia

Minimizing salivary gland radiation exposure is one effective strategy to minimize radiation-induced xerostomia. For example, in selected patients with early and moderate stages, well-lateralized oral and oropharyngeal carcinomas, ipsilateral irradiation treatment of the primary site, and ipsilateral neck spares salivary gland function on the uninvolved side, without compromising locoregional control.32

In conventional RT, reducing the mean dose to the contralateral SMG below 40 Gy is possible with a reasonable dose coverage.³³ Limiting the mean parotid dose to 31 Gy or less and mean minor salivary gland dose to 11 Gy or less in patients with lymphoma having RT to the H&N reduces the risk of xerostomia.³⁴

Using Positioning Devices, Shielding, and Conformational Field Planning

Positioning devices, shielding, and conformational field planning may also minimize salivary damage. The use of computed tomography (CT)-based delineation guidelines for organs at risk in the H&N should reduce inter- and intraobserver variability and therefore unambiguous reporting of possible dose-volume-effect relationships.³⁵

Intensity-modulated radiotherapy (IMRT) reduces doses when compared with standard three-dimensional conformal radiotherapy (CRT)^36,37 and reduces xerostomia.³⁸

Parotid gland sparing IMRT for patients with HNC improves xerostomia-related QoL compared with CRT both at rest and during meals: patients with laryngeal cancer had fewer complaints but benefited equally from IMRT compared with patients with oropharyngeal cancer.³⁹ IMRT in the treatment of nasopharyngeal carcinoma produced significant reductions in the occurrence rates and severity of acute skin reaction, neck fibrosis, trismus, and xerostomia.⁴⁰

SMG dose reduction to less than 39 Gy and without target underdosing is feasible in some patients at the expense of modestly higher doses to some other organs.41 Stimulated SMG flow rates decrease exponentially by 1.2% as mean doses increased up to 39 Gy threshold and then plateau near zero. At mean doses of 39 Gy or less, but not higher, flow rates recover over time at 2.2% a month. Similarly, the unstimulated salivary flow rates (USFRs) decrease exponentially by 3% as the mean dose increases and recovers over time if the mean dose was less than 39 Gy. IMRT replanning reduces mean contralateral SMG dose by an average of 12 Gy, achieving 39 Gy or less in five of eight patients, without target underdosing, and increasing the mean doses to the parotid glands and swallowing structures by an average of 2 to 3 Gy.

However, others have found that by 1 year after RT, normal tissue complication probability curves for IMRT and CRT were comparable with a median toxic dose (uniform dose leading to a 50% complication probability) of 38 and 40 Gy, respectively.⁴² Helical tomotherapy (TomoTherapy HiArt System; Accuracy, Sunnyvale, California, United States) of the parotid gland seems to largely preserve the function.⁴³

Using Radiation-Protective Agents

Radiation-protective agents can protect salivary glands in animal models, but translation of agents from animal testing to be used as prophylactic adjuncts or postexposure treatments in RT has been slow. Agents approved for the purpose by the U.S. Food and Drug Administration include amifostine (Ethyol, Ethiofos, WR-2721), an organic thiophosphate prodrug for alleviating xerostomia associated with RT. Amifostine is a free-radical scavenger, and it also accelerates DNA repair.⁴⁴ The selective protection of certain tissues of the body by amifostine is believed to be due to higher alkaline phosphatase activity, higher pH, and vascular permeation of normal tissues.

Acute xerostomia was significantly lessened in intravenous (IV) amifostine-treated patients with HNC (51%) compared with controls (78%).⁴⁵ Salivary output was significantly raised above that that for untreated controls 1 year after RT,⁴⁵ and xerostomia was reduced at 2-year follow-up.⁴⁶ Xerostomia after RT in a study by another group was similarly lower in amifostine-treated patients (57.5%) with HNC compared with controls (70%).⁴⁷ Amifostine was initially administered intravenously before chemotherapy or RT, but because of adverse effects and the cost of delivery, it is now provided through the subcutaneous (SC) route.⁴⁸ A study of 20 patients with HNC having RT-CT examined SC amifostine versus historical data of IV amifostine found outcomes from SC similar to those from IV but with reduced nausea/vomiting and hypotension after SC administration and showed xerostomia with SC drug as 42% (12 months) and 29% (18 months).⁴⁹ The Groupe d’Oncologie Radiothérapie Tête Et Cou study compared IV (200 mg/m²) versus SC (500 mg/d) amifostine in HNC and found less hypotension with SC amifostine,⁵⁰ which is virtually identical with other results.⁵¹

Both amifostine and IMRT are able to partially preserve parotid function after RT, although the effect of IMRT appears greater.⁵²

However, amifostine has not been shown to provide significant radioprotective effects on salivary glands in high-dose radioactive iodine–treated patients with differentiated thyroid cancer.⁵³

Pilocarpine (Salagen) is a slowly hydrolyzed muscarinic agonist with no nicotinic effects, which can increase secretion by the exocrine glands. The salivary, sweat, lacrimal, gastric, pancreatic, and intestinal glands and the mucous cells of the respiratory tract may be stimulated. In animal models, preirradiation treatment with pilocarpine induces a compensatory response at lower doses in the irradiated gland and at higher doses in the nonirradiated gland, reducing late damage, because of stimulation of unirradiated or surviving cells to divide.^54,55 A prospective, double-blind, placebo-controlled, randomized trial by the same authors in patients with HNC having RT showed that the concomitant administration of pilocarpine have no effect on parotid flow rate complications; however, patient-rated xerostomia scores showed trends toward less dryness-related complaints and there was reduced loss of parotid flow 1 year after RT in those patients who received pilocarpine and a mean parotid dose of more than 40 Gy.⁵⁶

In patients with HNC treated with bilateral RT in a double-blind, placebo-controlled, randomized trial using 5 mg of pilocarpine five times a day during RT, there was an improved overall QoL and less oral discomfort.⁵⁷ Others have also found some improvements,⁵⁸ although the clinical impact of the effect on salivation is not well defined.

The use of pilocarpine both during and after RT is also beneficial.⁵⁹

Repositioning of Surgical Salivary Gland

Surgical repositioning of the SMG out of the planned RT field to the submental space can protect the gland.⁶⁰ One study suggested that SMG transfer procedure is superior to pilocarpine in the management of RT-induced xerostomia.⁶¹

However, as RT is now delivered to the H&N with conformal fields/IMRT and tomotherapy, the reduction in the volume of high-dose radiation allows sparing of high-dose exposure to all salivary glands in many cases, with the increased potential for residual gland function and stimulation with sialogogues.

Management of the Patient with Cancer Liable to Hyposalivation

Recent recommendations for the management of patients with cancer liable to hyposalivation are as follows⁶²:

• Patients with cancer should be regularly assessed for salivary gland dysfunction (SGD);

• The management of SGD should be individualized;

Treatment is largely palliative and preventative in nature. Because oral dryness is a subjective complaint, it is not surprising that there is a great variation in the patient’s threshold of discomfort or other symptoms and it is also affected by tolerance and adaptation over time.63–65

Clinical Features

Oral complaints (often the presenting features) can include

• Xerostomia;

• Oral soreness or burning sensation;

• Difficulty in eating dry foods;

• Difficulty in speaking for long periods of time, the development of hoarseness, or there may be a clicking quality of the speech as the tongue tends to stick to the palate;

• Difficulty in swallowing;

• Difficulty in controlling dentures;

• Need of putting up a glass of water at night (and, sometimes, resulting nocturia); and

• Complications such as unpleasant taste or loss of sense of taste, oral malodor, caries, candidosis, and sialadenitis.

Common Terminology Criteria for Adverse Events (CTCAE v3.0) are as in Table 27.5.⁶⁷

A positive response to the questions in Table 27.6 is significantly associated with reduced salivary output.⁶⁸

Table 27.5 Common Terminology Criteria for Adverse Events

Table 27.6 Features of Hyposalivation: Questions and Responses

Chronic complications of hyposalivation may include the following (Table 27.7):

• Shift in the oral microflora and risk of oral infection (caries, candidosis, bacterial sialadenitis) (Figs. 27.1 to 27.4);

• Oral malodor;

• Altered/reduced taste;

• Mucosal dryness and sensitivity;

• Impaired chewing: patients with reduced or increased salivary flow, however, may not present measurable alterations in masticatory efficiency⁶⁹;

• Difficulty in swallowing;

• Difficulty in denture use and function, but there are few clinical research studies on the effect of hyposalivation on denture retention and mucosal trauma⁷⁰;

• Nutritional defects; and

• Altered speech.

Table 27.7 Clinical Symptoms of Xerostomia

Scales focused on xerostomia are shown in Table 27.8. The UW-QoL saliva domain seems to be a suitable means of screening for dry mouth in head-and-neck clinics and can be used to trigger interventions.⁷¹

Patient self-reported, rather than physician-assessed, scores should be the main end points in evaluating xerostomia because correlations between RTOG/EORTC grades and salivary flow rates are poor; in contrast, significant correlations are found between the patient self-reported scores and nonstimulated or stimulated salivary flow rates. No significant correlation was found between the RTOG/EORTC grades and the Xerostomia-Related Quality of Life Scale scores.^72–74

Table 27.8 Quality of Life Scales