Correspondence

Staging of Head and Neck Cancer

N Engl J Med 1995; 332:1787-1790June 29, 1995

Article

To the Editor:

Brennan et al. (Feb. 16 issue)1 are to be commended for their important and well-executed study. This is an excellent example of the use of emerging molecular-biology techniques to address important clinical questions. The clinical usefulness of the study would have been augmented, however, if the authors had provided details of the radiotherapy administered.

Because radiation kills cells exponentially, its success is inversely related to the log of the clonogenic cells within the treated volume.2 The tumor-cell burden is higher in patients with positive margins, and it is for this reason that radiation decreases the rate of local recurrence in patients with head and neck cancer by only 50 percent (from 60 percent to 30 percent) when given to patients with histologically positive margins, as compared with a decrease of 67 percent (from 30 percent to 10 percent) when given to patients at high risk for local recurrence but with histologically negative margins.3-5 The recurrence rate among patients with margins that were histologically negative but positive on analysis with the polymerase chain reaction (PCR) who were treated with resection and postoperative radiation in the study by Brennan et al. is similar to that among patients with histologically positive margins. This is not surprising, for in both situations residual cancer remains.

The degree of successful local control with resection that results in negative margins on PCR is undefined. This study suggests that local control in this subgroup of patients (which we presume received adjuvant irradiation) is excellent; there were no local failures. We believe that the results with local control with resection alone would be inferior. Despite the improvement in sensitivity with a PCR-based test, the limitations of sampling persist. Therefore, we will need to continue to assess the risk of local recurrence using all clinical and pathological data available (e.g., tumor and node stage, histologic grade, and presence or absence of perineural invasion).

The excellent local control achieved in this setting is also consistent with the fact that radiation is most successful when the residual number of clonogenic cells is lower. It appears that a finding of margins that are negative on PCR can be used to identify a subgroup of patients in whom local control can be consistently preserved with radiation. Overlooking the importance of this component of the treatment may jeopardize local control in the group in which success can be most assured.

In conclusion, all patients with locally advanced head and neck cancer are at risk for local recurrence after resection, whether they have histologically positive margins or histologically negative margins that are positive or negative on PCR. Postoperative radiation significantly decreases the rate of local recurrence in patients with locally advanced disease, and its biologic effectiveness increases from 50 percent (positive margins) to 100 percent (negative margins on PCR) as the tumor burden decreases. The absence of recurrence in the group with negative margins on PCR probably results from the successful combination of surgery and postoperative radiation.

Ronald D. Ennis, M.D.
Columbia University College of Physicians and Surgeons, New York, NY 10032

Jonathan P.S. Knisely, M.D.
Lynn D. Wilson, M.D., M.P.H.
Yale University School of Medicine, New Haven, CT 06510

5 References

1
Brennan JA, Mao L, Hruban RH, et al. Molecular assessment of histopathological staging in squamous-cell carcinoma of the head and neck. N Engl J Med 1995;332:429-435
Full Text | Web of Science | Medline

2
Fischer JJ. Radiobiology of therapeutic radiation. In: Ariyan S, ed. Cancer of the head and neck. St. Louis: C.V. Mosby, 1987:25-50.

3
Vikram B, Strong EW, Shah JP, Spiro R. Failure at distant sites following multimodality treatment for advanced head and neck cancer. Head Neck Surg 1984;6:730-733
CrossRef | Medline

4
Chen TY, Emrich LJ, Driscoll DL. The clinical significance of pathological findings in surgically resected margins of the primary tumor in head and neck carcinoma. Int J Radiat Oncol Biol Phys 1987;13:833-837
CrossRef | Web of Science | Medline

5
Amdur RJ, Parsons JT, Mendenhall WM, Million RR, Stringer SP, Cassisi NJ. Postoperative irradiation for squamous cell carcinoma of the head and neck: an analysis of treatment results and complications. Int J Radiat Oncol Biol Phys 1989;16:25-36
CrossRef | Web of Science | Medline

To the Editor:

In the study by Brennan et al., tumor cells were detected in histopathologically negative surgical margins and cervical lymph nodes of 25 patients with squamous-cell carcinoma of the head and neck through the identification of p53 gene mutations. This elegant approach is time consuming and limited by the fact that p53 mutations are present in only half of head and neck cancers.1

As an alternative approach we tried a simple immunocytochemical assay using monoclonal antibody A45-B/B3 to cytokeratins for the detection of individual squamous-cell-carcinoma cells in bone marrow.2 This method can detect 1 tumor cell in 1 million bone marrow cells and is therefore 100 times more sensitive than the molecular p53 assay used by Brennan et al. The specificity of this approach has been demonstrated by analysis of bone marrow from 75 control patients without carcinoma.2 Tumor cells in bone marrow were detected at frequencies of 1 to 207 positive cells per million marrow cells in 10 of 31 patients with cancer of the head and neck region (32 percent) (Table 1Table 1Incidence of Cytokeratin-Positive Tumor Cells in Bone Marrow of Patients with Squamous-Cell Carcinomas of the Oromaxillofacial Region.). Comparison with established risk factors revealed a positive correlation with the histologic grade of the primary tumor (P = 0.04) (Table 1). The presence of cytokeratin-positive cells in bone marrow was independent of the lymph-node involvement (Table 1), suggesting that these cells are direct descendants of the primary tumor.

A preliminary follow-up analysis showed that relapse occurred in 5 of the 10 patients (50 percent) with tumor cells in bone marrow, as compared with 6 of the 21 patients (29 percent) without such cells (length of observation, 4 to 23 months). Four bone marrow–positive patients (40 percent) died of cancer-related illness, whereas only two of the marrow-negative patients (10 percent) died during the observation period. Thus, although the main cause of death in patients with head and neck tumors is locoregional relapse, the presence of tumor cells in the bone marrow might be a manifestation of the tendency of an individual primary tumor to metastasize. In fact, the rate of bone marrow micrometastases is comparable to the incidence of distant metastases (26 percent) found in the autopsy study by Slootweg et al.,3 supporting the view that early hematogenous dissemination occurs more frequently than expected on the basis of clinicoradiologic tumor staging.

Klaus Pantel, M.D.
Universität München, 80336 Munich, Germany

Hans Gath, M.D.
Ernst Heissler, M.D.
Universitätsklinikum Rudolf-Virchow, 13353 Berlin, Germany

3 References

1
Boyle JO, Hakim J, Koch W, et al. The incidence of p53 mutations increases with progression of head and neck cancer. Cancer Res 1993;53:4477-4480
Web of Science | Medline

2
Pantel K, Schlimok G, Angstwurm M, et al. Methodological analysis of immunocytochemical screening for disseminated epithelial tumor cells in bone marrow. J Hematother 1994;3:165-173
CrossRef | Medline

3
Slootweg PJ, Bolle CW, Koole R, Hordijk GJ. Cause of death in squamous cell carcinoma of the head and neck: an autopsy study on 31 patients. J Craniomaxillofac Surg 1992;20:225-227
CrossRef | Web of Science | Medline

To the Editor:

The definitive treatment of patients with head and neck cancer has traditionally included radical surgery, radiation therapy, or a combination of the two.1 The choice is typically individualized according to the extent of the primary tumor, the extent of nodal involvement, and the possibility of organ preservation. The study by Brennan et al., in which surgical margins and resected lymph nodes were screened for specific p53 mutations present in the primary tumor, illustrates a new approach that may usefully incorporate molecular-biology techniques to improve care in these patients. However, we question whether the authors considered the possibility of “field cancerization” in some of the patients in whom p53 mutations were found. Extensive areas of a patient's aerodigestive tract may manifest gene mutations caused by a common environmental insult, such as tobacco smoke or ethanol.2,3 In these patients it may be impossible to obtain truly negative margins on molecular analysis without excessive surgical disfigurement or morbidity.

Furthermore, it would have been helpful to have complete data on the postoperative treatment provided and locoregional control for each patient. This information would help clinicians judge how useful these molecular studies would be over and above investigation of other clinical factors known to affect locoregional control, such as disease site, tumor–node–metastasis (TNM) staging, and tumor thickness. Preliminary evidence indicates that head and neck tumors with wild-type p53 and those with mutated p53 are equally sensitive to radiation therapy4,5; this suggests that p53 data may be less helpful as a stratification variable to determine which patients should receive postoperative radiation.

Lastly, it should be emphasized that the follow-up in these patients was short (median, 13 months) and that the pattern of recurrences may change with additional follow-up. Caution is in order if one applies the reported findings to daily medical practice; it is far from clear what the optimal approach should be.

Gary D. Kao, M.D.
Marc Rudoltz, M.D.
University of Pennsylvania, Philadelphia, PA 19104-6095

5 References

1
Snow GB, Clark JR, eds. Multimodality therapy for head and neck cancer. New York: Thieme Medical, 1992.

2
Field JK, Zoumpourlis V, Spandidos DA, Jones AS. p53 expression and mutations in squamous cell carcinoma of the head and neck: expression correlates with the patients' use of tobacco and alcohol. Cancer Detect Prev 1994;18:197-208
Web of Science | Medline

3
Chung KY, Mukhopadhyay T, Kim J, et al. Discordant p53 gene mutations in primary head and neck cancers and corresponding second primary cancers of the upper aerodigestive tract. Cancer Res 1993;53:1676-1683
Web of Science | Medline

4
Brachman DG, Beckett M, Graves D, Haraf D, Vokes E, Weichselbaum RR. p53 mutation does not correlate with radiosensitivity in 24 head and neck cancer cell lines. Cancer Res 1993;53:3667-3669
Web of Science | Medline

5
Slichenmyer WJ, Nelson WG, Slebos RJ, Kastan MB. Loss of a p53-associated G1 checkpoint does not decrease cell survival following DNA damage. Cancer Res 1993;53:4164-4168
Web of Science | Medline

To the Editor:

Brennan et al. suggest that a molecular search for mutated tumor cells (p53 gene) in the surgical margins of head and neck carcinomas is a powerful predictor of local tumor recurrence. They provide information on mutated p53 genes for all of their 25 patients but neglect to tell us in which patients the tumors recurred and which patients had recurrent or primary carcinoma at the beginning of the study. This information must have been available to them and is important for an assessment of a new staging test.

Furthermore, the authors do not indicate why the second examination of additional slides from the same patients, which apparently was not planned initially, was performed. Was it the finding that tumor cells made up 28 percent of the population in the histopathologically negative margin M5 of Patient 5? By not telling us which patients had recurrent tumors, they deprived us of a chance to correlate the percentage of tumor cells in the margins with the likelihood of recurrence.

To assess the sensitivity of this test, a correlation between the percentage of cells seen under the microscope (positive resection margins in a total of eight patients) and the percentage of tumor cells identified by molecular analysis would be useful. A comment about how the examination of resection margins under the microscope is normally conducted in their hospital and how that compares with the examinations performed in their study would be very informative.

Srdjan Denic, M.D.
Lewistown Hospital, Lewistown, PA 17044

Author/Editor Response

The authors reply:

To the Editor: As suggested by Dr. Denic, tumors with a larger number of neoplastic cells may be more likely to recur. The average percentage of neoplastic cells per margin in patients with recurrences ranged from 0.25 percent to 28 percent (Patients 3, 5, 13, 15, and 24). Tumors with positive molecular margins that did not recur contained an average of 0.05 percent to 10 percent neoplastic cells per margin. We analyzed 500 μm of tissue, whereas light microscopy can be used to assess only a small fraction of this amount (total, approximately 30 μm). Thus, estimates of neoplastic cells per given volume of tissue are not readily comparable. Patients with recurrent tumors at presentation were noted in Table 1 with an “R,” and we described the way in which resection margins routinely are analyzed and the way those under study were analyzed.

In response to Dr. Ennis and colleagues: all our patients received a standard dose of radiation therapy (approximately 60 Gy). Patients with a large number of neoplastic cells (e.g., the five patients in whom large numbers of such cells were readily identified by standard light microscopy) had persistent disease with rapid recurrence despite full-dose radiation therapy. As noted above, patients with a lower number of neoplastic cells per margin were less likely to have recurrences. we agree that the absence of recurrence in the group with negative margins on PCR may well be due to a combination of both surgery and postoperative radiation therapy. However, it would be of interest to identify a subgroup of patients who might be spared adjuvant irradiation.

Dr. Pantel and colleagues propose using a simple immunocytochemical assay. The advantages of modern molecular diagnostics are the substantial sensitivity of PCR and the specificity of certain mutations associated with the clonal proliferation of neoplastic cells. Immunohistochemical analysis, by contrast, often depends on the identification of proteins that occur in both neoplastic and non-neoplastic cells. Our own analysis of lymph nodes suggests little value in cytokeratin staining as compared with molecular analysis. We do not readily accept the premise that standard immunohistochemical analysis is either more sensitive or more specific than molecular analysis.

Drs. Kao and Rudoltz provide a long list of drawbacks to our study. We have previously cast considerable doubt on the concept of field cancerization in bladder cancer and certainly considered the possibility in this study.1 Drs. Kao and Rudoltz question the specificity of the identification of the same mutation as was present in the primary tumor. We detected the same neoplastic clones that were present in the primary tumor and do not offer an assessment based on equivocal staining patterns of individual cells. Our own data and those of others point out that a second primary cancer often contains an unrelated and distinct p53 mutation.2,3 Using a similar approach in a study of colorectal cancer, investigators detected the same K-ras and p53 mutations in lymph nodes that were initially identified in the primary tumor.4 Moreover, even if a neoplastic clone is at an earlier stage of evolution (i.e., potentially preinvasive and devoid of some of the genetic alterations present in the primary tumor), the presence of this clone may still indicate a substantial risk of continued genetic evolution and recurrence. Tumors from all of our patients contained a p53 mutation, so sensitivity to radiotherapy based on p53 status was not a factor.5 However, provocative studies in animals have suggested that p53 status may determine the response to ionizing radiation. Finally, with a median follow-up of 20 months, there have been no additional recurrences.

Although caution is in order, the daily failures of our approach to cancer in humans make it mandatory to consider new approaches. There are precious few examples of molecular approaches that have been successfully transferred from the bench to the bedside. It is already apparent that our molecular approach is much more sensitive than standard histopathological techniques. Obviously, the accepted approach for patients with head and neck cancer will be derived from prospective testing in ongoing controlled clinical trials. However, the ability to augment current staging by molecular analysis already adds an important new dimension for consideration. If stricken with cancer, which staging approach would our detractors opt for?

Joseph A. Brennan, M.D.
Li Mao, M.D.
David Sidransky, M.D.
Johns Hopkins University, Baltimore, MD 21205

5 References

1
Sidransky D, Frost P, Von Eschenbach A, Oyasu R, Preisinger AC, Vogelstein B. Clonal origin of bladder cancer. N Engl J Med 1992;326:737-740
Full Text | Web of Science | Medline

2
Chung KY, Mukhopadhyay T, Kim J, et al. Discordant p53 gene mutations in primary head and neck cancers and corresponding second primary cancers of the upper aerodigestive tract. Cancer Res 1993;53:1676-1683
Web of Science | Medline

3
Koch WM, Boyle JO, Mao L, Hakim J, Hruban RH, Sidransky D. p53 gene mutations as markers of tumor spread in synchronous oral cancers. Arch Otolaryngol Head Neck Surg 1994;120:943-947
Web of Science | Medline

4
Hayashi N, Arakawa H, Nagase H, et al. Genetic diagnosis identifies occult lymph node metastases undetectable by the histopathological method. Cancer Res 1994;54:3853-3856
Web of Science | Medline

5
Lowe SW, Bodis S, McClatchey A, et al. p53 status and the efficacy of cancer therapy in vivo. Science 1994;266:807-810
CrossRef | Web of Science | Medline

Citing Articles (2)

Citing Articles

1
Alexander Bertalanffy, Alexander Kober, Petra Bertalanffy, Burkhard Gustorff, Odette Gore, Sharam Adel, Klaus Hoerauf. (2005) Transcutaneous Electrical Nerve Stimulation Reduces Acute Low Back Pain during Emergency Transport. Academic Emergency Medicine 12:7, 607-611
CrossRef
2
Martin Nuhr, Klaus Hoerauf, Alexander Bertalanffy, Petra Bertalanffy, Nathalie Frickey, Carmen Gore, Burkhard Gustorff, Alexander Kober. (2004) Active Warming During Emergency Transport Relieves Acute Low Back Pain. Spine 29:14, 1499-1503
CrossRef