Symptoms in early head and neck cancer: An inadequate indicator

Symptoms in early head and neck cancer: An inadequate indicator ROBERT W. DOLAN, MD, FACS,* CHARLES W. VAUGHAN, MD, FACS, and NABIL FULEIHAN, MD, Boston, Massachusetts

Screening programs show promise in increasing the rate of early detection of head and neck cancers in high-risk populations. Prout et al (Otolaryngol Head Neck Surg 1997;116:201-8) examined the usefulness of a large-scale screening program for head and neck cancer in an inner city population by primary care physicians. Symptom assessment was based on the American Cancer Society’s “Seven Warning Signs for Cancer,” (Cancer manual. 8th ed. Boston: American Cancer Society, Massachusetts Division; 1990. p. 40-64) 4 of which are relevant to the head and neck. However, these signs may be insufficient for detection of early head and neck cancer. We analyzed these and other typical symptoms to determine their role in early detection. Coincident medical problems, tobacco abuse, and alcohol abuse were also analyzed. Our findings indicate that no symptom or symptom complex is strongly correlated with early head and neck cancer for any subsite except the glottis. Symptom duration is an unreliable indicator of the duration of disease. However, patients under medical supervision are more likely to have their cancers detected early, supporting the value of surveillance by the primary care physician. The absence of definite early warning signs for most head and neck cancers suggests the need to develop essential screening criteria. Defining the population that is at high risk for head and neck cancer and subjecting it to an aggressive screening protocol is essential. (Otolaryngol Head Neck Surg 1998;119:463-7.)

Detecting head and neck (H/N) cancers at an early stage is believed to be the most effective means to reduce death and disfigurement from this disease. There

were 42,400 newly diagnosed H/N squamous cell cases and 11,500 deaths in the United States in 1993.1 Despite advances in treatment during the last 30 years, this mortality rate has remained stable. However, screening programs show promise in increasing the rate of early detection in high-risk populations.2.3 Prout et al4 examined the usefulness of a large-scale H/N cancer screening program by primary care physicians in an inner city population. They used risk and symptom assessment and visual inspection for oral mucosal lesions. Risk assessment involved selecting patients for screening who were 40 years or older and had a history of tobacco use. Symptom assessment was based on the American Cancer Society’s “Seven Warning Signs for Cancer,”5 4 of which are relevant to the H/N: (1) a sore that does not heal, (2) a lump, (3) difficulty swallowing, and (4) persistent hoarseness. Their results showed that directed surveillance in this setting was effective in generating referrals to otolaryngologists for patients with early cancers. However, the American Cancer Society’s warning signs may be insufficient for symptom assessment in early H/N cancer, except for persistent hoarseness. A nonhealing sore (persistent ulcer) is easily visible in the oral cavity only, a lump or neck mass is a late symptom, and the association of dysphagia with early H/N cancer is undocumented. The most common symptoms of H/N cancer are local pain, voice change, weight loss, referred pain, neck mass, bleeding, stridor, and dysphagia.3,6 To determine which, if any, are indicators of early H/N cancer, we analyzed these symptoms with respect to occurrence, duration, and cancer stage, overall, and with respect to the anatomic subsite of the cancer. The association of cancer stage with coincident medical problems, tobacco abuse, and alcohol abuse was also determined. METHODS

From the Department of Otolaryngology–Head and Neck Surgery, Boston Medical Center. *Dr. Dolan is currently at the Department of Otolaryngology–Head and Neck Surgery, University of Oklahoma Health Sciences Center. Reprint requests: Charles W. Vaughan, MD, FACS, Department of Veterans Affairs Medical Center, Department of OtolaryngologyHead and Neck Surgery, 150 S Huntington Ave, Boston, MA 02130 Copyright © 1998 by the American Academy of Otolaryngology– Head and Neck Surgery Foundation, Inc. 0194-5998/98/$5.00 + 0 23/1/87345

The study set consisted of data that had been collected and recorded in standardized form by otolaryngology–H/N surgical residents and staff for all patients with squamous cell carcinoma involving the mucosal surfaces of the H/N (excluding sinonasal malignancies) admitted to the otolaryngology H/N service at the Boston Veterans Affairs Medical Center from July 1988 through June 1995. The data included the presenting symptoms and their duration; a complete history and physical examination including any history of prior cancers; a 463

464

Otolaryngology– Head and Neck Surgery November 1998

DOLAN et al.

Table 1. Symptoms associated with H/N cancer Signs and symptoms

Local pain Neck mass Voice change/hoarseness Dysphagia Weight loss Referred pain Bleeding Asymptomatic Stridor Cranial nerve dysfunction

Present at time of diagnosis (%)

53 46 44 38 29 26 8 8 6 1

history of social/substance abuse; the location and size of the index cancer as determined by physical examination, diagnostic radiology, including plain film, CT, and MRI examinations; and endoscopy and biopsy with the patient under general anesthesia, as had been described on standardized forms, which included tumor maps, and staged in accordance with the staging system of the American Joint Committee for Cancer (AJCC, 1988).5 The oral cavity was further divided into anterior and posterior subsites with the second molar and Stensen’s duct forming the plane separating the two. As part of the history, medical conditions were recorded. To be included and counted, the patient must have been taking medication related to the condition to establish a link between the presence of the medical condition and the existence of regular periodic medical supervision. To confirm and validate the association of early staged cancers with reduced mortality for our data set, we analyzed survival information from the tumor registry at the Veterans Affairs Medical Center. Inferential statistics were done using χ2 analysis. Terminology criteria are as follows—asymptomatic: no symptoms, incidental finding; bleeding: presence of blood or blood-tinged sputum; cranial nerve dysfunction: sensory or a motor deficit of recent onset (prior 6 months) not caused by stroke or other non-cancer-related disease; early: cancer is stage II or less, according to AJCC criteria (eg, T1 or T2 and without metastasis); local pain: report of dull or sharp pain in the area of the cancer; medical conditions: other illnesses for which the patient had been receiving treatment; neck mass: any mass of the neck, detected by physical examination and/or CT/MRI, thought to represent a cervical metastasis; referred pain: attribution of pain arising away from the cancer as being caused by the cancer; stridor: noisy breathing from upper airway obstruction; T: refers to the stage of the primary; T1/2: includes T1 and T2 cancers; T3/4: includes T3 and T4 cancers (the AJCC may be referenced for further information regarding the meaning of T for each subsite); voice change: any change from the usual voice as noted by the patient or any vocal sound degradation as noted by the physician, including

changes in resonance and articulation such as a “hot potato” voice or a muffled voice; and weight loss: unintentional loss of weight of greater than 10 lb during the prior 6 months. FINDINGS

Data from 492 patients were reviewed. There were 459 men, and the average age for all patients was 63 years (range 32 to 88 years). All patients had abused tobacco (76% active and 24% former), and 74% had abused alcohol (68% active and 6% former). Anatomic sites of the index cancer included the anterior oral cavity (108), posterior oral cavity (35), retromolar trigone (14), oropharynx (81), nasopharynx (12), hypopharynx (67), supraglottis (58), glottis (93), subglottis (4), and unknown primary with neck mass (20). Medical conditions included chronic obstructive pulmonary disease, chronic heart failure, hypertension, coronary artery disease, chronic renal failure, active liver cirrhosis, peptic ulcer disease, seizure disorder, rheumatoid arthritis, diabetes mellitus, schizophrenia, and chronic hematologic disorders. All records were complete with regard to the recording of symptoms. The most common symptoms associated with H/N cancer were local pain, neck mass, voice change, dysphagia, weight loss, referred pain, bleeding, stridor, and cranial nerve dysfunction (Table 1, and by subsites in Table 2). The most common combinations of 2 symptoms associated with each subsite are in Table 3. The duration of symptoms had been recorded in 484 (98%) records. No significant differences were found in overall duration of symptoms, or according to stage, alcohol use, or coexistent medical conditions. The overall duration of symptoms was similar across all subsites at 3.5 months. Duration according to stage was also consistent at between 3.4 and 3.6 months. The average in tobacco-only users was 3.54 months, and the average in tobacco and alcohol abusers was 3.47 months. The average duration in patients with no coexistent medical conditions was 3.51 months; a single medical condition, 3.42 months; and multiple medical conditions, 3.51 months. Prevalence of symptoms according to stage is shown in Table 4, and by stage and subsite in Table 5. Survival according to stage is as follows (excluded are patients with evidence of disease and those dead of other causes): T1/2, 23% dead of disease (22-month median follow-up) and 40% alive and free of disease (20-month median follow-up); and T3/4, 36% dead of disease (20month median follow-up) and 25% alive and free disease (20-month median follow-up). Patients with no coexistent medical conditions requiring regular medical follow-up had more advanced staged cancers compared to those under

Otolaryngology– Head and Neck Surgery Volume 119 Number 5

DOLAN et al.

465

Table 2. Symptoms associated with each subsite (%)

Anterior oral cavity Posterior oral cavity Retromolar trigone Oropharynx Nasopharynx Hypopharynx Supraglottis Glottis Subglottis

Local pain

Voice Weight change Dysphagia loss

70.5 69.2 70.6 67.9 41.7 52.2 48.3 18.3 25.0

18.1 26.5 7.7 27.2 25.0 50.7 63.8 90.3 100

27.6 41.2 23.1 49.4 33.3 61.2 60.3 18.3 50.0

32.4 23.5 30.8 28.4 33.3 43.3 41.4 11.8 25.0

medical supervision for a single coexistent medical condition (P = 0.09), and for multiple medical conditions (P = 0.05). There was no significant difference in cancer stage between patients with single versus multiple medical conditions (Table 6). Patients with a history of alcohol abuse presented with later staged cancers (P = 0.01) (Table 7), the presence of coexistent medical conditions not significantly different between tobacco users versus tobacco and alcohol users. DISCUSSION

To determine which, if any, symptoms may aid early diagnosis of H/N cancer, symptoms associated with this cancer were evaluated for occurrence, duration, and stage, overall, and with respect to the anatomic subsite of the cancer. With the exception of voice change and glottic cancer, no symptom or symptom complex was found to be strongly associated with, or a reliable indicator of, early cancer. The symptoms tended to be relatively nonspecific, becoming more prevalent as the cancer progressed. Bleeding, stridor, and cranial nerve dysfunction were found primarily in late-stage cancers. Referred pain also was a late sign but was found slightly more often as an early sign in cancers of the retromolar trigone and hypopharynx. Unintentional weight loss usually was present in advanced disease. The symptoms related to the H/N in the American Cancer Society’s “Seven Warning Signs for Cancer”5 appear to be primarily associated with late disease. “A lump” (neck mass) is present in 16.7% to 56.7% of T1/2 primary cancers of the H/N (excluding glottic cancers); these cancers are not early, and their prognosis is poor. The term “a sore that does not heal” is difficult to evaluate because it encompasses 2 possible symptoms: persistent ulcer and/or persistent pain. A persistent ulcer is not a common symptom of patients with H/N cancer. An ulcer may be detected incidentally at physical examination,7 and this is represented in our data set as “asymptomatic,” most commonly in the anterior oral

Referred pain

Neck mass

Bleeding

Asymptomatic

24.8 35.3 38.5 37.0 16.7 38.8 20.7 11.8 25.0

40.0 41.2 15.4 54.3 50.0 68.7 44.8 18.3 50.0

4.8 0.0 7.7 11.1 16.7 13.4 13.8 2.2 0.0

15.2 8.8 23.1 14.8 0.0 4.5 3.4 1.1 0.0

Cranial nerve Stridor dysfunction

1.0 2.9 7.7 2.5 0.0 11.9 12.1 11.8 25.0

1.0 0.0 0.0 0.0 8.3 0.0 0.0 0.0 0.0

Table 3. The most common combinations of 2 symptoms associated with each subsite Sites

Anterior oral cavity Posterior oral cavity Retromolar trigone Oropharynx Hypopharynx Supraglottis

Symptoms/signs

Local pain & neck mass Local pain & dysphagia Local pain & weight loss Local & referred pain Local pain & dysphagia Dysphagia & neck mass Voice change & dysphagia

%

31.4 35.3 30.8 30.8 43.2 43.3 37.9

Table 4. Prevalence of symptoms according to staging for all subsites Prevalence (%) Symptoms

Local pain Voice change Dysphagia Weight loss Referred pain Neck mass Bleeding Asymptomatic Stridor Cranial nerve dysfunction

T1/2

T3-4

37.8 3.8 21.8 16.0 12.6 23.1 3.4 14.3 3.4 0.0

66.9 50.4 54.2 39.4 38.1 58.1 11.9 2.1 9.3 1.0

cavity and oropharynx. However, if a sore that does not heal represents a persistent soreness (local pain), it is the most common symptom of H/N cancer. Dysphagia is an early symptom in hypopharyngeal cancer only (60%), closely followed by neck mass (56.7%) in this subsite. “Persistent hoarseness” (voice change) is the most prevalent early symptom in glottic cancer, but it is of little use in early diagnosis for the other subsites. Symptom duration with respect to subsite and stage was remarkably consistent at about 3.5 months and not

466

Otolaryngology– Head and Neck Surgery November 1998

DOLAN et al.

Table 5. Staging versus symptoms associated with subsite

n Local pain Voice change Dysphagia Weight loss Referred pain Neck mass Bleeding Asymptomatic Stridor Cranial nerve dysfunction

Anterior oral cavity

Posterior oral cavity

Retromolar trigone

T1/2

T3/4

T1/2

T3/4

T1/2

T3/4

T1/2

T3/4

T1/2

T3/4

T1/2

T3/4

T1/2

T3/4

T1/2

T3/4

52 56.3 6.3 12.5 16.7 8.3 22.9 2.1 31.3 0.0 0.0

56 84.9 30.2 43.4 45.3 35.8 52.8 7.5 1.9 1.9 1.9

19 61.1 11.1 22.2 11.1 11.1 27.8 0.0 16.7 5.6 0.0

16 86.7 46.7 60.0 40.0 60.0 53.3 0.0 0.0 0.0 0.0

7 33.3 16.7 16.7 0.0 33.3 16.7 0.0 50.0 0.0 0.0

7 100 0.0 28.6 57.1 42.9 14.3 14.3 0.0 14.3 0.0

30 53.6 14.3 21.4 17.9 17.9 39.3 0.0 28.6 3.6 0.0

51 76.0 32.0 64.0 32.0 46.0 62.0 16.0 6.0 2.0 0.0

5 40.0 0.0 40.0 20.0 0.0 40.0 20.0 0.0 0.0 20.0

7 42.9 42.9 28.6 42.9 28.6 57.1 14.3 0.0 0.0 0.0

30 36.7 30.0 60.0 30.0 30.0 56.7 6.7 6.7 6.7 0.0

37 65.7 68.6 62.9 51.4 45.7 77.1 20.0 2.9 17.1 0.0

21 57.9 47.4 42.1 31.6 10.5 21.1 10.5 10.5 0.0 0.0

37 43.2 75.7 70.3 45.9 24.3 54.1 16.2 0.0 16.2 0.0

67 13.6 90.9 9.1 10.6 9.1 4.5 1.5 1.5 4.5 0.0

26 30.8 88.5 42.3 15.4 19.2 53.8 3.8 0.0 26.9 0.0

Oropharynx Nasopharynx Hypopharynx Supraglottis

Table 6. Stage at presentation and chronic medical problems Medical problems

None

Single

T1/2 T3/4

61 (43%) 81 (57%)

164 (51%) 155 (49%)

Multiple

94 (54%) 80 (46%)

Table 7. Stage at presentation and tobacco versus tobacco and alcohol abuse

T1/2 T3/4

Tobacco alone

Tobacco and alcohol

69 (59%) 48 (41%)

156 (45%) 187 (55%)

affected by use of tobacco, alcohol, or the presence of chronic medical conditions. In this and other studies,8 there is no difference in symptom duration between early- and late-stage cancers. This finding is illogical; nevertheless, it has led some to conclude that early diagnosis may not make an impact on prognosis. However, we conclude that recall of symptom duration by our patient population is unreliable and not useful as an indicator for duration of disease. The association of cancer stage with concurrent illness and alcohol/tobacco abuse was also analyzed. We found a positive correlation between the presence of chronic medical conditions under treatment and early detection, suggesting that patients under medical supervision are more likely to have their cancers detected early. This finding supports the value of surveillance by the primary care physician in the early detection of H/N cancer. Patients who abused alcohol tended to have later

Glottis

stage cancers at presentation. The reasons for this finding are not discernible from the present data set. These cancers may be biologically more aggressive, the host defenses may be impaired, or these patients are less likely to seek adequate medical care. SUMMARY

No symptom or symptom complex is strongly correlated with early H/N cancer for any subsite except the glottis. Local pain, voice change, neck mass, dysphagia, weight loss, referred pain, bleeding, stridor, and cranial nerve dysfunction make up a useful symptom set; however, it encompasses predominantly late symptoms. Local pain and voice change, although present relatively often in early cancers, are found more commonly in late-stage disease (except voice change in glottic cancers). Symptom duration is an unreliable indicator of the duration of disease. The absence of definite early warning signs for most H/N cancers suggests the need to develop essential screening criteria. Defining the population at high risk for developing H/N cancer and subjecting it to an aggressive screening protocol is essential. It is through prospective and outcome studies that we will be able to determine whether such steps will make a difference in unmasking the silence of most early H/N cancers. We acknowledge the assistance of Herbert Kayne, PhD (Department of Epidemiology and Biostatistics, Boston University School of Public Health), for performing the statistical analysis. REFERENCES 1. Spitz MR. Epidemiology and risk factors for head and neck cancer. Semin Oncol 1994;21:281-8.

Otolaryngology– Head and Neck Surgery Volume 119 Number 5

2. Amsel Z, Engstrom PF, Strawitz JG. The dentist as a referral source of first episode head and neck cancer patients. J Am Dent Assoc 1983;106:195-7. 3. Ford CN, Rebstock KE, Brandenburg JH. Lessons from the network and next steps: the Wisconsin head and neck cancer network. Adv Cancer Control: Res and Dev 1983:279-90. 4. Prout MN, Sidari JN, Witzburg RA, et al. Head and neck cancer screening among 4611 tobacco users older than forty years. Otolaryngol Head Neck Surg 1997;116:201-8.

DOLAN et al.

467

5. Cancer manual. 8th ed. Boston: American Cancer Society, Massachusetts Division; 1990. p. 40-64. 6. Collins SL. Avoiding delay and misdiagnosis of head and neck cancer: rare tumors with common symptoms. Compr Ther 1995; 21:59-67. 7. Thompson LW. Head and neck cancer: early detection. Semin Surg Oncol 1989;5:168-75. 8. Vernham GA, Crowther JA. Head and neck carcinoma—stage at presentation. Clin Otolaryngol 1994;19:120-4.

Your comprehensive reference to valuable research information...

O

tolaryngology–Head and Neck Surgery provides its readers with timely information on recent treatments and advances in otolaryngology.

YES!

Please send me the Five-Year Cumulative Index to Otolaryngology–Head and Neck Surgery (1991-1995). I’ve indicated my method of payment below. All prices include shipping. ❏ $45.00 U.S

❏ $52.96 Canada*

❏ $49.50 International

* Includes Canadian GST.

Now you can have easy access to that information with the five-year cumulative author and subject index to Volumes 104-113 of Otolaryngology–Head and Neck Surgery. The Cumulative Index is your comprehensive reference guide, including a list of every article in the journal from 1991 through 1995. All articles are indexed both by subject and by author. The Cumulative Index features: •Offset and bold Subject Headings •Subject Entries containing complete article title, author(s), year of publication, volume, and page •Author Entries listing author(s), article title, author-toauthor referral, year of publication, volume, and page Order the Otolaryngology–Head and Neck Surgery Cumulative Index today!

Payment must accompany your order. ❏ Check enclosed (U.S. funds, payable to Otolaryngology– Head and Neck Surgery) ❏ MasterCard ❏ VISA ❏ Discover Credit Card # _______________________ Exp. Date ________ Signature ___________________________________________ Name ______________________________________________ Institution ___________________________________________ Address_____________________________________________ City ______________________________ State ____________ ZIP ___________________ Country _____________________

☎ To order, just call 1-800-453-4351 Outside the U.S.: 314-453-4351 Fax: 314-432-1158

Or return coupon to: Mosby, Inc. Journal Subscription Services 11830 Westline Industrial Dr. St. Louis, MO 63146 U.S.A.