Annals of Surgical Oncology Sign the Guestbook
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by El-Tamer, M.
Right arrow Articles by Braverman, A. S.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by El-Tamer, M.
Right arrow Articles by Braverman, A. S.
Annals of Surgical Oncology 9:340-345 (2002)
© 2002 Society of Surgical Oncology

ORIGINAL ARTICLES

Prognoses of T4 Breast Cancer Subsets

Mahmoud El-Tamer, MD, Sabir Hussain, MD, Jeremy Weedon, PhD, Haji Chalchal, MD, Ashis Chakrabarti, MD, Chul Sohn, MD and Albert S. Braverman, MD

From the Department of Surgery (ME-T), Columbia Presbyterian Medical Center, New York, New York; Dr. Everett Chalmers Regional Hospital (SH), Oncology Department, Fredericton, New Brunswick, Canada; Department of Scientific Computing (JW), Division of Hematology, Department of Medicine (AC, ASB), and Department of Radiation Oncology (CS), Health Sciences Center at Brooklyn, State University of New York, New York; and Allan Blair Cancer Center (HC), Department of Medical Oncology, Regina, Saskatchewan, Canada.

Correspondence: Address correspondence and reprint requests to: Albert S. Braverman, MD, Box 55, Heath Sciences Center at Brooklyn, State University of New York, 450 Clarkson Ave., Brooklyn, NY 11203-2098; Fax: 718-270-1544; E-mail: abraverman{at}netmail.hscbklyn.edu


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 PATIENTS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Background: The relative prognoses of subsets of patients with T4 breast tumors, since the advent of neoadjuvant therapy, are unknown, although inflammatory carcinoma (T4d) is considered to have the worst prognosis.

Methods: Subsets of T4 patients were analyzed for the incidence of distant metastases at presentation (M1; n = 263). T4M0 patients treated with neoadjuvant therapy (n = 126) were analyzed for relapse-free survival (RFS). T4d tumors with (Cut/CW+) and without (Cut/CW-) skin nodules, posterior fixation, or both were analyzed separately.

Results: Fewer patients with T4d (Cut/CW-) tumors had distant metastases at presentation than T4d (Cut/CW+) patients or T4b and T4c patients (P = .001, .001, and .009, respectively). RFS was longer for T4b patients than for T4c patients (P = .018) or T4d (Cut/CW+) patients (P = .003). RFS of the T4d (Cut/CW+) patients was shorter than for T4d (Cut/CW-) patients (P = .050).

Conclusions: The incidence of distant metastases at presentation was lowest, and RFS was longest, for patients with T4d tumors not grossly involving the skin or posterior structures. Patients whose tumors grossly invaded both skin and posterior structures (T4c) or those with T4d tumors grossly invading either most frequently presented with distant metastases and had the shortest RFS.

Key Words: Breast cancer • Inflammatory carcinoma • T4 breast tumor • Prognosis


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
PATIENTS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Primary breast tumors are designated T4a if they invade the chest wall, T4b if they invade the skin, T4c if both are invaded, and T4d if they are inflammatory, with diffuse, brawny induration of the skin and an erysipeloid edge.1,2 Patients with T4 tumors of any type, without metastases, are staged as IIIB (T4, any N, M0), regardless of axillary nodal status,1 but T4d tumors are considered to have the worst prognosis.2–5 There is extensive literature on T4d tumors, but the few studies on the incidence and prognoses of the other T4 subsets6,7 predate the systematic use of neoadjuvant chemotherapy.

Although often purely diffuse (Cut/CW-), inflammatory carcinomas may form nonulcerated gross cutaneous (satellite) nodules and/or discrete masses fixed posteriorly to the chest wall (Cut/CW+). Both are designated T4d; the prognostic significance of these local invasive changes has not been investigated.

To determine the percentage of patients presenting with each type of T4 tumor who were or were not potentially curable, we have recorded the incidence of distant metastases at presentation (stage IV; T4, any N, M1) in each T4 subset of a series of 263 patients. To determine the prognosis of each subset, we present data on the duration of relapse-free survival (RFS) in a series of 126 T4M0 patients who were treated with neoadjuvant chemotherapy, surgery, and radiation. We also present data on the incidence of distant metastases at presentation and RFS in a series of 112 patients presenting with T3 tumors (one diameter >=5 cm without cutaneous involvement or posterior fixation)1 who received similar treatment.


   PATIENTS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 PATIENTS AND METHODS
RESULTS
 DISCUSSION
 REFERENCES
 
Data on 263 patients who presented to the Kings County Medical Center Breast Clinic from 1980 to 2000 with T4 tumors are the basis of this analysis. We also include data on 112 patients with T3 tumors. All patients presenting without distant metastases were offered primary systemic (neoadjuvant) therapy, followed by mastectomy or tumorectomy and radiation.8

There are ambiguities in the definitions of the T4 subsets. Tumors in which inflammation is localized and due to ulceration of discrete, gross cutaneous masses should probably be designated T4b, rather than T4d. Of our 99 patients without diffuse inflammation, but with localized cutaneous involvement (T4b), 44 presented with local ulceration.

Fixation to the chest wall is the criterion for T4a and T4c tumors, 1 but only surgery can reliably determine whether tumors fixed posteriorly on physical examination invade the pectoralis muscle or the chest wall itself. Because many patients with posteriorly fixed tumors present with distant metastases, surgery is often not performed. When it is, the extent of invasion may have been altered by neoadjuvant chemotherapy. We therefore designated tumors fixed posteriorly on physical examination as T4a or T4c; the posterior location of such tumors was confirmed by mammography and sonography.

Our criteria for inflammatory carcinoma (T4d) were the simultaneous presence of diffuse edema (peau d’ orange), erythema, warmth, and tenderness. We did not distinguish patients with and without discrete, palpable masses. We did not consider demonstration of microscopic permeation of the dermal lymphatics a necessary or a sufficient condition for the T4d designation.2,9 Cutaneous inflammation may be due to permeation of the deep breast lymphatics alone.10–12 Several categories of T4d tumors have been proposed: occult inflammatory carcinoma refers to tumors without inflammation or gross skin involvement, in which microscopic dermal lymphatic invasion is demonstrated pathologically.12 However, it is not possible to predict whether such cutaneous infiltration would, if untreated, result in discrete skin nodules (T4b) or diffuse inflammation (T4d). Some T4d tumors have been categorized as secondary when a significant interval elapsed between presentation with a breast mass and the development of cutaneous inflammation.13 This distinction presupposes extended clinical observation of patients with breast cancers, but neoadjuvant therapy is now used immediately and is often followed by mastectomy and radiation. Patients’ histories are not reliable enough to determine the interval, if any, between the development of a breast mass and the first signs of inflammation.14

Some patients (n = 30) with diffuse inflammation of the breast also presented with satellite breast skin nodules or discrete masses that were fixed posteriorly. The American Joint Committee on Cancer staging criteria were construed to mean that such tumors should be designated T4d, rather than T4a or T4b, implying that inflammation takes precedence over all other criteria for definition of T4 subsets. Inflammation has not, however, been shown to imply a worse prognosis than gross invasion of local structures. To determine whether satellite nodules or posterior fixation associated with T4d tumors implies a worse prognosis than inflammation alone, we separately analyzed subsets of T4d tumors with (Cut/CW+) or without (Cut/CW-) satellite nodules, posterior fixation, or both.

Supraclavicular nodal involvement at presentation was determined by physical examination and confirmed by needle aspiration in doubtful cases. We did not consider such patients as presenting with distant metastases (M1) if no other metastases were found by staging studies. From 1982 to 1988, such patients were officially staged as IIIB,15 and recent data support this designation.16 There were 44 patients with supraclavicular nodal involvement, 20 of whom also presented with distant metastases.

Nodal status was defined by clinical criteria as N0 or >N0; the latter category included N1 and N2 patients and also those with supraclavicular nodal involvement. Clinical nodal status was confirmed in 142 of the 149 patients who had mastectomies or tumorectomies. Seven other patients were upstaged from N0 to N1 by surgery.

At presentation all patients were staged with chest films, bone scans, and, in most cases, hepatic computed tomography or radionuclide scanning, especially in those with hepatomegaly or abnormal liver function studies. Questionable pulmonary parenchymal or mediastinal involvement was confirmed or eliminated with chest computed tomography scanning. The results of these studies were used to determine the incidence of distant metastases at presentation.

To determine the duration of RFS of the subsets of T4 patients without metastases, we analyzed only those patients who had received neoadjuvant therapy and whose response to that therapy was recorded. Of the 171 T4 patients presenting without metastases, 45 refused neoadjuvant therapy or did not report for more than one course of treatment, and most of these were lost to follow-up. Thus, 126 T4 patients were available for analysis of RFS duration. Compliance with therapeutic recommendations was good.17

After or during neoadjuvant therapy, almost all stage III patients and some with involved supraclavicular nodes only were offered mastectomy or, in some cases, tumorectomy, followed in all cases by radiation. The exceptions were certain older patients with medical illnesses, patients with supraclavicular nodes whose breast tumors were unresponsive to neoadjuvant therapy, and those who relapsed before surgery could be performed. Of the patients presenting without distant metastases, 54 (89%) of 61 T3 patients and 103 (82%) of 126 T4 patients had mastectomies or tumorectomies, followed by radiation. The breasts and adnexa of all M0 patients who did not have surgery were also irradiated.

Relapse was defined as palpable local disease on physical examination or definite evidence of distant metastases on imaging studies. The latter studies were not repeated after presentation unless the patient had specific symptoms or signs, such as cough, dyspnea, hepatomegaly, abnormal liver function studies, or pain. Like others, we found that intervals of 6 to 12 months often elapsed between the appearance of bone metastases on radionuclide scanning and the first symptoms of pain. Radiologically apparent osteoblastic metastases were also often painless for many months. Nodular pulmonary metastases were also often asymptomatic for long periods.18– 25

RFS was defined as the interval between the date of presentation and the last visit at which the patient was free of evidence of symptoms or signs of relapse. To avoid overestimating the duration of RFS in patients who did not return until long after symptoms appeared, the following procedure was used: when >6 months had elapsed between the visit at which relapse was discovered and the previous visit, the date of relapse onset was estimated from the nature of the relapse. Patients were considered to have had objective evidence of local or distal recurrence at least 3 months before they presented with demonstrable relapse. Bone metastases were presumed to have been detectable on bone scanning 12 months before presentation with pain.

Comparison of prevalence of distant metastases among groups was formalized by using Fisher’s exact test. Comparison of RFS among groups was performed with Kaplan-Meier plots and the log-rank test. All statistical analyses were conducted with SAS software (SAS Institute, Inc., Cary, NC).


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 PATIENTS AND METHODS
 RESULTS
DISCUSSION
 REFERENCES
 
Distant Metastases at Presentation
The prevalence of distant metastases at presentation for patients staged with chest films, bone scans, and hepatic scans is listed in Table 1. Twenty (18%) of the 112 T3 patients and 92 (35%) of the 263 T4 patients had metastases at presentation. The difference was significant (P = .024). Significantly fewer T3 patients had metastases at presentation than each of the T4 subsets, with the exception of the T4a and T4d (Cut/CW-) groups. Among the T4 patients, significantly fewer of the T4d (Cut/CW-) subset had distant metastases at presentation than the T4b (P = .001), T4c (P = .001), or T4d (Cut/CW+; P = .009) subsets.


View this table:
[in this window]
[in a new window]
  		TABLE 1. Incidence of metastases at presentation, other than supraclavicular node involvement, in 112 T3 patients and subsets of 263 patients with T4 tumors
 
Relapse-Free Survival
Of patients without distant metastases at presentation, except supraclavicular node involvement, 61 T3 and 126 T4 patients were available for RFS analysis. All received neoadjuvant therapy. Of the T3 patients, 46 received cyclophosphamide, doxorubicin, and 5-flurouracil (CAF); 7 received cyclophosphamide, methotrexate, and 5-flurouracil (CMF); 3 received cyclophosphamide and paclitaxel (CyTx); and 5 received tamoxifen. Of the T4 patients, 104 received CAF, 8 received CMF, 9 received CyTx, and 5 received tamoxifen. The total number of local and distant relapses over a 12-year follow-up period is listed in Table 2.


View this table:
[in this window]
[in a new window]
  		TABLE 2. Local and distant relapses in T3 and T4 patients,n (%)
 
Kaplan-Meier plots for the RFS of the T3 patients and subsets of T4 patients are presented in Fig. 1. The median months of RFS for each T4 subset and the percentage relapse-free at 2 and 5 years are listed in Table 3. RFS was significantly longer for T3 than for T4 patients (P = .037). T4b patients had a significantly longer RFS than T4c (P = .018) or T4d (Cut/CW+) patients (P = .003). RFS was significantly longer for T4d (Cut/CW-) than for T4d (Cut/CW+) patients (P = .050). At 5 years, a significant difference between the RFS of T3 and T4 patients persisted.



View larger version (16K):
[in this window]
[in a new window]
  		FIG. 1. Kaplan-Meier plots of relapse-free survival durations of patients with T3 breast tumors and those with each T4 subset. Patients with inflammatory carcinomas (T4d) with (Cut/CW-) and without (Cut/CW-) satellite nodules, posterior fixation, or both were analyzed separately.

 

View this table:
[in this window]
[in a new window]
  		TABLE 3. Median months relapse free (RF) and percentage RF at 2 and 5 y in 61 T3 patients and subsets of 126 T4 patients without metastases at presentation who received neoadjuvant therapy
 
Among the patients with T4 tumors, we distinguished two groups, depending on whether their tumors displayed one or more than one criterion for the T4 designation. Thus, group 1 included tumors that invaded only posterior structures or skin or displayed inflammation uncomplicated by gross invasion of the skin or posterior structures: T4a, T4b, and T4d (Cut/CW-). Group 2 included tumors that invaded both posterior structures and skin or inflammatory tumors associated with gross invasion of either: T4c and T4d (Cut/CW+). When compared with both of these groups of T4 patients, the RFS of the T3 patients was also significantly longer (P = .0004). The group 1 T4 patients had a significantly longer RFS than the group 2 T4 patients (P = .001; Table 4).


View this table:
[in this window]
[in a new window]
  		TABLE 4. Percentage relapse-free (RF) survival at 5 y of 61 T3 patients and grouped subsets 126 T4 patients without metastases at presentation who received neoadjuvant therapy
 
The relation of nodal status (N0 or >N0) to the incidence of relapse in the T3 and the T4 groups was determined by a log-rank test. Data on nodal status were available for 60 of 61 T3M0 patients and 121 of 126 T4M0 patients. Of the 60 T3 patients, only 2 of 16 N0 patients relapsed, as compared with 25 of 44 of the >N0 patients (P = .023). During the first 4 years of follow-up, relapse seemed to occur more frequently among the >N0 than the N0T4 patients, but there was no significant overall difference by log-rank test (P = .135). Of 121 T4 patients, 16 of 29 N0 patients relapsed after a median interval of 35 months. Of 92 >N0T4 patients, 59 relapsed after a median interval of 18 months.


   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 PATIENTS AND METHODS
 RESULTS
 DISCUSSION
REFERENCES
 
Breast cancers with cutaneous inflammation (T4d) have been considered to be "the most lethal of all breast tumors,"2 but there have been few systematic comparisons between the prognoses of all T4 subsets, and some of these predated contemporary neoadjuvant therapy.6,7 We have therefore analyzed the prognostic implications of both inflammation and gross invasion of adjacent structures. To do so, we distinguished inflammatory carcinomas associated with satellite nodules or posterior fixation from those with neither, although we designated both groups as T4d. Because it is difficult to determine which structures are invaded by tumors fixed posteriorly on physical examination, we considered posterior fixation a sufficient criterion for designating tumors as T4a or T4c. Duration of RFS was determined only for those M0 patients who received neoadjuvant therapy.

Gross invasion of the skin, posterior structures, or both implied a worse prognosis than inflammation alone. Patients with inflammatory carcinomas (T4d) uncomplicated by gross local invasion had the lowest incidence of metastases at presentation, and those presenting without metastases had the longest RFS. Conversely, patients whose tumors grossly invaded both skin and posterior structures (T4c) and those with inflammatory carcinomas that grossly invaded either had the highest incidence of metastases at presentation; those presenting without metastases had the shortest RFS. Gross invasion of the skin or posterior structures may, in many cases, be due to delay in seeking medical care for a breast tumor, rather than to rapid tumor growth. Such neglect of primary tumors may also be responsible for the increased incidence of metastases at presentation and the shorter RFS of patients with T4c tumors and those inflammatory carcinomas that have produced satellite nodules, invaded posterior structures, or both. The distant micrometastases arising from such neglected tumors may be larger and less responsive to systemic therapy than those from uncomplicated inflammatory carcinoma

There was no significant difference between the RFS duration of T4M0 patients with and without nodal involvement, but there was a significant difference between these groups of T3 patients. This is consistent with American Joint Committee on Cancer staging criteria and suggests that gross invasion of adjacent structures is a risk factor for the development of distant metastases, independent of nodal involvement.

Our results suggest that the prognosis of breast cancer patients with cutaneous inflammation alone may be superior to that of patients whose tumors grossly invade adjacent structures. Patients with T4d tumors uncomplicated by such invasion are less likely to present with distant metastases and have a longer RFS after neoadjuvant chemotherapy and local treatment than patients with other types of T4 tumors.

Received for publication May 29, 2001. Accepted for publication December 17, 2001.


   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 PATIENTS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 

  1. Fleming ID, Cooper JS, Henson DE, et al. American Joint Committee on Cancer Staging Manual. 5th ed. Philadelphia: Lippincott-Raven, 1998.
  2. Jaiyesimi IA, Buzdar AU, Hortobagyi G. Inflammatory carcinoma of the breast. J Clin Oncol 1992; 10: 1014–24.[Abstract]
  3. Paradiso A, Tommasi S, Brandi M, et al. Cell kinetics and hormonal receptor status in inflammatory breast carcinoma. Cancer 1989; 64: 1922–7.[CrossRef][Medline]
  4. Droulias CA, Sewell CW, McSweeney MB, Powell RW. Inflammatory carcinoma of the breast: a correlation of clinical, radiologic and pathologic findings. Ann Surg 1976; 184: 217–22.[Medline]
  5. Garcia I, Dietrich PY, Aapro M, Vauthier G, Vadas L, Engel E. Genetic alterations of c-myc, c-erbB-2, and c-Ha-ras protooncogenes and clinical associations in human breast carcinomas. Cancer Res 1989; 49: 6675–9.[Abstract/Free Full Text]
  6. Rubens RD, Armitage P, Winter PJ, Tong D, Hayward JL. Prognosis in inoperable stage III carcinoma of the Breast. Eur J Cancer 1977; 13: 805–11.
  7. Sutherland CM, Mather FJ. Long-term survival and prognostic factors in patients with regional breast cancer (skin, muscle and/or chest wall involvement). Cancer 1985; 55: 1389–97.[CrossRef][Medline]
  8. Morrow M, Braverman A, Thelmo W, et al. Multimodal therapy for locally advanced breast cancer. Arch Surg 1986; 121: 1291–6.[Abstract/Free Full Text]
  9. Barker JL, Nelson AJ, Montague ED. Inflammatory carcinoma of the breast. Radiology 1976; 121: 173–6.[Abstract]
  10. Rosen PP. Invasive mammary carcinoma.In: Harris JR, Lippman ME, Morrow M, Hellman S, eds. Diseases of the Breast. New York: Lippincott-Raven, 1996: 393–444.
  11. Ellis DL, Teitelbaum SL. Inflammatory carcinoma of the breast. Cancer 1974; 33: 1045–7.[CrossRef][Medline]
  12. Lucas FV, Perez-Mesa C. Inflammatory carcinoma of the breast. Cancer 1978; 41: 1595–605.[CrossRef][Medline]
  13. Henderson M, McBride C. Secondary inflammatory breast cancer. South Med J 1988; 81: 1512–7.[Medline]
  14. Bloom HJG. The influence of delay on the natural history and prognosis of breast cancer. Br J Cancer 1965; 19: 228–62.
  15. Beahrs OH, Myers MH. Manual for Staging of Cancer. 2nd ed. Philadelphia: Lippincott, 1983: 129–30.
  16. Brito RA, Valero V, Buzdar AU, et al. Long-term results of combined-modality therapy for locally advanced breast cancer with ipsilateral superclavicular metastases. J Clin Oncol 2001; 19: 628–33.[Abstract/Free Full Text]
  17. Berger D, Braverman A, Sohn CK, Morrow M. Patient compliance with aggressive multimodal therapy in locally advanced breast cancer. Cancer 1988; 61: 1453–6.[CrossRef][Medline]
  18. Tamin R, Donegan WL. Screening for recurrent breast cancer. J Clin Oncol 1987; 5: 62–7.[Abstract]
  19. Brown ML. Bone scintigraphy in benign and malignant tumors. Radiol Clin North Am 1993; 31: 731–8.[Medline]
  20. Front D, Schneck SO, Frankel A, Robinson E. Bone metastases and bone pain in breast cancer. JAMA 1979; 242: 1747–8.[Abstract/Free Full Text]
  21. Schutte HE. The influence of bone pain on the results of bone scan. Cancer 1979; 44: 2039–42.[CrossRef][Medline]
  22. Chaudary MA, Maisy MN, Shaw PJ, Rubens RD, Hayward JL. Sequential bone scans and chest radiographs in the post-operative management of early breast cancer. Br J Surg 1983; 70: 517–8.[Medline]
  23. Khansur T, Haick A, Patel B, Balducci L, Vance R, Thigpen T. Evaluation of bone scan as a screening work-up in primary and local-regional recurrence of breast cancer. Am J Clin Oncol 1987; 10: 167–70.[Medline]
  24. Joo KG, Parthasarathy KL, Bakshi SP, Rosner D. Bone scintigrams: their clinical usefulness in patients with breast carcinoma. Oncology 1979; 36: 94–8.[Medline]
  25. Podoloff DA. Malignant bone disease.In: Henkin RE, Boles MA, Dillehay GL, et al., eds. Nuclear Medicine.Vol. 2. St. Louis: Mosby, 1996: 1208–22.



This article has been cited by other articles:


Home page
 Ann OncolHome page
U. Guth, G. Singer, A. Schotzau, I. Langer, H. Dieterich, C. Rochlitz, L. Herberich, W. Holzgreve, and E. Wight
Scope and significance of non-uniform classification practices in breast cancer with non-inflammatory skin involvement: a clinicopathologic study and an international survey
Ann. Onc., October 1, 2005; 16(10): 1618 - 1623.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by El-Tamer, M.
Right arrow Articles by Braverman, A. S.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by El-Tamer, M.
Right arrow Articles by Braverman, A. S.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS