Barry W. Brown

Prevention of Second Primary Tumors with Isotretinoin in Squamous-Cell Carcinoma of the Head and Neck

BACKGROUND Patients with head-and-neck cancers who are free of disease after local therapy remain at high risk for both recurrent and second primary tumors. Retinoids have proved efficacious in the treatment of premalignant oral lesions and are promising agents for the prevention of epithelial carcinogenesis. METHODS We prospectively studied 103 patients who were disease-free after primary treatment for squamous-cell cancers of the larynx, pharynx, or oral cavity. After completion of surgery or radiotherapy (or both), these patients were randomly assigned to receive either isotretinoin (13-cis-retinoic acid) (50 to 100 mg per square meter of body-surface area per day) or placebo, to be taken daily for 12 months. RESULTS There were no significant differences between the two groups in the number of local, regional, or distant recurrences of the primary cancers. However, the isotretinoin group had significantly fewer second primary tumors. After a median follow-up of 32 months, only 2 patients (4 percent) in the isotretinoin group had second primary tumors, as compared with 12 (24 percent) in the placebo group (P = 0.005). Multiple second primary tumors occurred in four patients, all of whom were in the placebo group. Of the 14 second cancers, 13 (93 percent) occurred in the head and neck, esophagus, or lung. CONCLUSIONS Daily treatment with high doses of isotretinoin is effective in preventing second primary tumors in patients who have been treated for squamous-cell carcinoma of the head and neck, although it does not prevent recurrences of the original tumor.

RANDOMIZED TRIAL ADDRESSING RISK FEATURES AND TIME FACTORS OF SURGERY PLUS RADIOTHERAPY IN ADVANCED HEAD-AND-NECK CANCER

Abstract Purpose: A multi-institutional, prospective, randomized trial was undertaken in patients with advanced head-and-neck squamous cell carcinoma to address ( 1 ) the validity of using pathologic risk features, established from a previous study, to determine the need for, and dose of, postoperative radiotherapy (PORT); ( 2 ) the impact of accelerating PORT using a concomitant boost schedule; and ( 3 ) the importance of the overall combined treatment duration on the treatment outcome. Methods and Materials: Of 288 consecutive patients with advanced disease registered preoperatively, 213 fulfilled the trial criteria and went on to receive therapy predicated on a set of pathologic risk features: no PORT for the low-risk group ( n = 31); 57.6 Gy during 6.5 weeks for the intermediate-risk group ( n = 31); and, by random assignment, 63 Gy during 5 weeks ( n = 76) or 7 weeks ( n = 75) for the high-risk group. Patients were irradiated with standard techniques appropriate to the site of disease and likely areas of spread. The study end points were locoregional control (LRC), survival, and morbidity. Results: Patients with low or intermediate risks had significantly higher LRC and survival rates than those with high-risk features ( p = 0.003 and p = 0.0001, respectively), despite receiving no PORT or lower dose PORT, respectively. For high-risk patients, a trend toward higher LRC and survival rates was noted when PORT was delivered in 5 rather than 7 weeks. A prolonged interval between surgery and PORT in the 7-week schedule was associated with significantly lower LRC ( p = 0.03) and survival ( p = 0.01) rates. Consequently, the cumulative duration of combined therapy had a significant impact on the LRC ( p = 0.005) and survival ( p = 0.03) rates. A 2-week reduction in the PORT duration by using the concomitant boost technique did not increase the late treatment toxicity. Conclusions: This Phase III trial established the power of risk assessment using pathologic features in determining the need for, and dose of, PORT in patients with advanced head-and-neck squamous cell cancer in a prospective, multi-institutional setting. It also revealed the impact of the overall treatment time in the combination of surgery and PORT on the outcome in high-risk patients and showed that PORT acceleration without a reduction in dose by a concomitant boost regimen did not increase the late complication rate. These findings emphasize the importance of coordinated interdisciplinary care in the delivery of combined surgery and RT.

Evaluation of the dose for postoperative radiation therapy of head and neck cancer: first report of a prospective randomized trial.

PURPOSE This study was designed to determine in a prospective randomized trial the optimal dose of conventionally fractionated postoperative radiotherapy for advanced head and neck cancer in relation to clinical and pathologic risk factors. METHODS AND MATERIALS Between January 1983 and March 1991, 302 patients were enrolled on the study. This analysis is based on the first 240 patients entered through September 1989, of whom 221 (92%) had AJC Stage III or IV cancers of the oral cavity, oropharynx, hypopharynx, or larynx. The patients were stratified by postulated risk factors and randomized to one of three dose levels ranging between 52.2 Gy and 68.4 Gy, all given in daily doses of 1.8 Gy. Patients receiving > 57.6 Gy had a field reduction at this dose level such that boosts were only given to sites of increased risk. RESULTS The overall crude and actuarial 2-year local-regional recurrence rates were 25.4% and 26%, respectively. Patients who received a dose of < or = 54 Gy had a significantly higher primary failure rate than those receiving > or = 57.6 Gy (p = 0.02). No significant dose response could be demonstrated above 57.6 Gy except for patients with extracapsular nodal disease in the neck in whom the recurrence rate was significantly higher at 57.6 Gy than at > or = 63 Gy. Analysis of prognostic factors predictive of local-regional recurrence showed that the only variable of independent significance was extracapsular nodal disease. However, clusters of two or more of the following risk factors were associated with a progressively increased risk of recurrence: oral cavity primary, mucosal margins close or positive, nerve invasion, > or = 2 positive lymph nodes, largest node > 3 cm, treatment delay greater than 6 weeks, and Zubrod performance status > or = 2. Moderate to severe complications of combined treatment occurred in 7.1% of patients; these were more frequent in patients who received > or = 63 Gy. CONCLUSION With daily fractions of 1.7 Gy, a minimum tumor dose of 57.6 Gy to the whole operative bed should be delivered with a boost of 63 Gy being given to sites of increased risk, especially regions of the neck where extracapsular nodal disease is present. Treatment should be started as soon as possible after surgery. Dose escalation above 63 Gy at 1.8 Gy per day does not appear to improve the therapeutic ratio.

Phase I/II study of stereotactic body radiotherapy for spinal metastasis and its pattern of failure

OBJECT The authors report data concerning the safety, effectiveness, and patterns of failure obtained in a Phase I/II study of stereotactic body radiotherapy (SBRT) for spinal metastatic tumors. METHODS Sixty-three cancer patients underwent near-simultaneous computed tomography-guided SBRT. Spinal magnetic resonance imaging was conducted at baseline and at each follow-up visit. The National Cancer Institute Common Toxicity Criteria 2.0 assessments were used to evaluate toxicity. RESULTS The median tumor volume of 74 spinal metastatic lesions was 37.4 cm3 (range 1.6-358 cm3). No neuropathy or myelopathy was observed during a median follow-up period of 21.3 months (range 0.9-49.6 months). The actuarial 1-year tumor progression-free incidence was 84% for all tumors. Pattern-of-failure analysis showed two primary mechanisms of failure: 1) recurrence in the bone adjacent to the site of previous treatment, and 2) recurrence in the epidural space adjacent to the spinal cord. Grade 3 or 4 toxicities were limited to acute Grade 3 nausea, vomiting, and diarrhea (one case); Grade 3 dysphagia and trismus (one case); and Grade 3 noncardiac chest pain (one case). There was no subacute or late Grade 3 or 4 toxicity. CONCLUSIONS Analysis of the data obtained in the present study supports the safety and effectiveness of SBRT in cases of spinal metastatic cancer. The authors consider it prudent to routinely treat the pedicles and posterior elements using a wide bone margin posterior to the diseased vertebrae because of the possible direct extension into these structures. For patients without a history of radiotherapy, more liberal spinal cord dose constraints than those used in this study could be applied to help reduce failures in the epidural space.

Concomitant boost radiotherapy schedules in the treatment of carcinoma of the oropharynx and nasopharynx

Concomitant boost schedules are characterized by delivering the boost (10-12 fractions) as second daily treatments during rather than following the basic wide field irradiations. This results in shortening the overall time to administer 69-72 Gy from 7 1/2-8 weeks to 6 weeks, which we hoped would improve the tumor control rate by reducing the opportunity for tumor clonogens to regenerate during treatment. From August 1985 to August 1988, 79 patients with T2-4 carcinomas of the oropharynx (72 patients) or nasopharynx (7 patients) were treated according to 1 of the 3 variants of the concomitant boost technique. The median age of patients was 60 years (range: 19-84 years) and the male-to-female ratio was 2.6. The overall 2-year actuarial primary and nodal control rates by radiotherapy alone were 74% and 76%, respectively. The ultimate 2-year control rates after surgical salvage were 82% and 84%, respectively. If the boost given during the last 2-2 1/2 weeks of basic treatment, a slightly better primary control rate (p = 0.11) resulted than if the boost was delivered during the first 2-2 1/2 weeks or twice a week throughout the basic treatment. The 2-year actuarial primary control rate of the 13 patients receiving induction chemotherapy prior to radiotherapy was significantly lower than that of patients treated with radiation only (81% vs 34%, p = 0.01), but this could be partly attributed to a more advanced stage in the chemotherapy group. The acute mucosal reactions were, as expected, more severe than those observed with conventional fractionation. Fifty patients developed confluent mucositis covering more than half of the boost area. Such reactions lasted for more than 6 weeks in seven patients. Late complications, however, so far observed, have been few. Three patients experienced chronic mucosal tenderness, 1 chronic mucosal ulceration, 2 transient bone exposure, and 1 carotid rupture following salvage surgery. The results so far appear to be better than the outcome of conventional radiotherapy. Its real value will be determined in a prospective randomized study.

Expression of Blood-Group Antigen A — A Favorable Prognostic Factor in Non-Small-Cell Lung Cancer

BACKGROUND New prognostic factors are needed to guide the treatment of patients with non-small-cell lung cancer. We evaluated the prognostic value of altered expression of ABH blood-group antigens, which has been implicated in the multistep process of carcinogenesis and tumor progression. METHODS The presence of blood-group antigens was assessed immunohistochemically in paraffin-embedded tumor samples from 164 patients who underwent curative surgery for non-small-cell lung cancer from 1980 through 1982. Monoclonal antibodies were used to detect the A and B antigens, and Ulex europaeus agglutinin I to detect H antigen. RESULTS Survival of the 28 patients with blood type A or AB who had primary tumors negative for blood-group antigen A was significantly shorter than that of the 43 patients with antigen A-positive tumors (P less than 0.001) and of the 93 patients with blood type B or O (P = 0.002). The respective median survival times were 15, 71, and 39 months. Disease progressed significantly earlier in the 28 patients with tumors negative for blood-group antigen A than in the antigen A-positive patients (P less than 0.001). Expression of blood-group antigen B or H in tumor cells did not correlate with survival. Cox proportional-hazards regression analysis showed that expression of blood-group antigen A in tumor cells added significantly to the prediction of overall survival provided by other known prognostic factors among the patients with blood type A or AB (P = 0.004). CONCLUSIONS Expression of blood-group antigen A in tumor cells is an important favorable prognostic factor in patients with non-small-cell lung cancer. This variable needs to be considered in the design of future trials of therapy.

Impact of cyclophosphamide on long-term reduction in sperm count in men treated with combination chemotherapy for ewing and soft tissue sarcomas

Background. Treatment of cancer with multiple‐drug chemotherapy regimens or radiation therapy can cause either temporary azoospermia of various durations or permanent azoospermia in young men.

Methods correcting for multiple testing: operating characteristics

We examine the operating characteristics of 17 methods for correcting p-values for multiple testing on synthetic data with known statistical properties. These methods are derived p-values only and not the raw data. With the test cases, we systematically varied the number of p-values, the proportion of false null hypotheses, the probability that a false null hypothesis would result in a p-value less than 5 per cent and the degree of correlation between p-values. We examined the effect of each of these factors on family-wise and false negative error rates and compared the false negative error rates of methods with an acceptable family-wise error. Only four methods were not bettered in this comparison. Unfortunately, however, a uniformly best method of those examined does not exist. A suggested strategy for examining corrections uses a succession of methods that are increasingly lax in family-wise error. A computer program for these corrections is available.

Estimation of the screening error rate from the observed detection rates in repeated cervical cytology

Abstract A curious stepping-down in case rates was noticed in some studies where repeated screening was used to detect dysplasia, carcinoma in situ, and invasive cancer. The third and fourth observed apparent incidence rates were successively smaller than the initial observed incidence rate which is, of course, the second screening rate. Since the false-negative error rate in cervical screening is known to be high, it was suspected that this error, as it was expressed in successive screenings, was causing the step-down effect. A model designed to test this hypothesis gave results that fully explained the data. Error rates were calculated by least-squares fit of the model to the observed data, and false-negative error rates of 40 per cent for dysplasia, 20, 27 and 45 per cent for CIS, and 24 per cent for IC were found. The consequences of error rates of this magnitude are discussed.

The natural history of resectable metastatic melanoma (Stage IVA Melanoma)

One‐hundred‐two patients with malignant melanoma who had distant metastases surgically resected and were judged to be clinically free of disease (M. D. Anderson Stage IVA melanoma) were studied. The median survival for all the patients from time of diagnosis of stage IVA disease was 18 months. The site of the resected metastases did not appear to influence survival, being approximately the same for the brain (15 months), lung (16 months), intraabdominal (18 months), and skin and/or lymph nodes (23 months). The site of the resected metastases also did not influence the median disease‐free interval. Patients who had metastases resected from several organs at the same time had a median survival of 15 months, which was similar to that of patients with one resected site. Patients who were rendered Stage IVA on several occasions by surgical excisions had a median survival of 36 months. Thirty‐five patients received surgery only and 67 patients received adjuvant chemotherapy, immunotherapy, or combined chemoimmunotherapy after surgery. For the group treated with surgery only, the median disease‐free interval and survival from diagnosis of stage IVA disease were 6 months and 16 months, respectively, and for the adjuvant group 6 months and 21 months, respectively. Specifically, by the type of adjuvant therapy, the median disease‐free interval and survival from stage IVA for 23 patients receiving Corynebacterium parvum were 6.9 and 19 months; for 39 patients receiving BCG, eight months and 26 months; for 24 patients receiving BCG + DTIC, eight and 17.4 months; and for all 51 DTIC treated patients 6.3 and 17.8 months, respectively. Patients receiving BCG had a median survival superior to the surgery only group (P = 0.02). An increase in survival was seen predominantly in patients who achieved IVA status more than once and received BCG. Patients with recurrent soft‐tissue metastases appeared to benefit most from BCG in prolonging the disease‐free interval. Only 1/10 treated by surgery alone had a disease‐free interval longer than 1 year, compared with 9/16 who received BCG (P = 0.01). Stage IVA melanoma appears to be distinctly different in prognosis from Stage IVB melanoma and should be classified separately. Patients with recurrent soft‐tissue disease may benefit significantly from treatment with BCG.