---

Abstract

Background: Tumor infiltrating lymphocytes (TIL) are associated with spontaneous regression and improved survival in patients with melanoma. The aim of our study was to evaluate if histopathological detection of TIL is a reliable indicator to predict aggressive tumor biology.
 
Methods: We retrospectively analyzed 241 patients in a single institution from 2009-2011. The presence or absence of TIL was compared to preoperative white blood cell counts (WBC), age, tumor depth, mitotic rates, sentinel lymph nodes (SLN), and pathologic stage.
 
Results: Of 241 patients, 53% were male. Age was categorized as 65 years, with 55% being under 65 (mean=61). Patients >65 withlower WBC counts were more likely to have absence of TIL (p=0.03). Thick melanomas >4 mm were seen in 15% of our population. Patients with thinner melanomas had TIL present 48% of the time, compared to only 22% with thicker tumors (p=0.02). Of patients with positive SLNs, TIL was identified in 46% vs. 38% with a negative SLN (p=0.38); a mitotic rate >4 mm2 was identified in 31% vs. 21%, respectively (p<0.001). Preoperative mitotic rates were compared to overall stage, and the presence of low mitotic rates correlated to a lower stage (p<0.001). Preoperative TIL was compared to overall stage, and the presence of TIL did not correlate with an earlier staged melanoma (p=0.29). Multivariate analysis comparing clinical stage, TIL, and mitotic rate was performed, and the combination of TIL and degree of mitoses did not predict earlier staged melanoma when compared to mitotic rate alone (p=0.39).
 
Conclusion: Known factors such as positive SLNs, mitotic rates, and depth of invasion serve as reliable prognostic indicators independently. In our series, we found the presence of TIL to correlate with tumors <4 mm. When comparing TIL with mitosis, SLN, and early staged melanomas, there was no statistical significance to suggest benefit as a reliable prognostic indicator.

---

Keywords:

Melanoma; Tumor infiltrating lymphocytes; Prognosis

---

Introduction

Despite public awareness warning against increased sun exposure, cutaneous melanoma continues to be one of the leading causes of death of all the skin cancers. Both incidence and mortality rates have increased steadily worldwide [1]. Significant advances in adjuvant therapies have been made in the past ten years. One of the earliest publications documenting prognostic factors in patients with melanoma was published thirty five years ago, and tumor thickness, along with anatomical site, were felt to be dominant factors [2]. Additional factors included: pigmentation, lymphocyte infiltration, growth pattern and regression; however, these factors were found to have either indirect or no influence on survival. In 2001, the American Joint Committee on Cancer proposed major revisions on the staging system, and with the aid of a multi-institutional effort, several key prognostic indicators for melanoma were agreed upon; tumor thickness, ulceration, presence of positive lymph nodes, and visceral versus non visceral metastatic disease [3]. Accurately predicting prognosis aids in guiding appropriate management, determining the need for further investigation, and to assign risk status for patients that may benefit from closer surveillance, adjuvant therapy, or consideration for a clinical trial [1].The presence of tumor infiltrating lymphocytes (TIL) and their roles as a prognostic indicator for malignant melanomas has been a topic of much controversy. There currently is no consensus regarding the presences of TIL in melanoma; nonetheless, the general trend suggests that tumor infiltrating lymphocytes portend an added survival benefit. Tumor infiltrating lymphocytes are histopathological markers of the host immune response to melanoma [4]. Several studies have been able to correlate the presence of TIL with improved overall survival [1,4-6], however others have found no such relationship [7,8]. There have been numerous prognostic indicators that are known to play a role in melanoma, and depth of invasion has been one of the most sensitive factors [2]. As advances have been made, the presence of ulceration, mitosis, and sentinel lymph node positivity have added even more information on outcomes [3]. As much controversy continues to exist regarding tumor infiltrating lymphocytes as a potential mediator in prognosis, we aimed to evaluate if TILs correlated with prognosis in patients with malignant melanoma.

---

Methods

Using a search of our pathology database, we identified females diagnosed with PSCCB at our institution from 2007-2012. Inclusion criteria included female patients over the age of 18 years old with a primary tumor consisting of greater than 90% malignant cells of squamous origin. Likewise, exclusion criteria included squamous cell cancer that had metastasized to the breast from elsewhere or lesions that were not independent of the overlying skin or nipple. We then performed a retrospective review to evaluate patient characteristics, presentation, tumor characteristics, treatment modalities and outcomes. A literature review was also performed using the PubMed database using the key words “primary squamous cell cancer of the breast”, “primary squamous cell carcinoma of the breast” and “primary squamous cell breast cancer”.

---

Results

The first patient is a 52 year old healthy, non-smoking, postmenopausal, female with no personal or family history of breast, ovarian, or colon cancer who presented with a progressively enlarging left breast mass that she noticed one month prior to presentation. On examination of her left breast, a 6 centimeter mobile breast mass at the twelve o’clock position was noted with no skin changes, nipple discharge or lymphadenopathy. She underwent a bilateral diagnostic mammogram with a 2.5 centimeter dense mass noted in the left upper outer quadrant of her breast. An ultrasound guided core needle biopsy was performed that demonstrated a moderate to well differentiated keratinizing squamous cell carcinoma. Furthermore, a bilateral breast MRI was performed that depicted a 3.2 centimeter spiculated mass in the left upper outer quadrant at the two o’clock position with no evidence of lymph node, chest wall or right breast involvement. She underwent a left partial mastectomy with axillary sentinel lymph node biopsy. Final pathology demonstrated a 4.5 centimeter, stage pT2N0M0, moderately differentiated squamous cell carcinoma with negative lymph nodes and margins. ER/PR/Her2-neu were all negative with a Ki-67 proliferation index of 80%. Postoperatively, she was treated with adjuvant whole breast radiation and no systemic therapy. At her 28 month follow up, she continues to be disease free.The second patient is a 35 year old healthy, non-smoking, premenoapausal female with no personal or family history of breast, ovarian, or colon cancer who presented with a right breast mass that she palpated on a breast self-exam. On examination of the right breast, a 2 centimeter mass at the one o’clock position was noted with a negative axillary exam. She had a diagnostic mammogram performed demonstrating a 2.6 centimeter mass in the upper inner quadrant of her right breast adjacent to the chest wall. An ultrasound guided core needle biopsy was performed that demonstrated moderately differentiated keratinizing squamous cell carcinoma. A pre-operative PET scan was negative for metastasis. The patient underwent a bilateral skin sparing mastectomy with bilateral axillary sentinel lymph node biopsy and immediate reconstruction. Pathology demonstrated a right breast 1 centimeter, pT1bN0M0 moderately differentiated squamous cell carcinoma with negative lymph nodes and margins. In addition, ER/PR/Her2-neu receptors were negative with a Ki-67 proliferation index of 40%. Post-operatively, she received no adjuvant radiation or systemic therapy. At her eighteen month follow up, she continues to be disease free.

---

Discussion

Because PSCCB is so rare, prognosis and optimal treatment are still controversial [1-5,7-12]. There is a small body of research regarding this topic which consists mainly of isolated case reports and a few larger case series. One of the larger series by Hennessey et al. includes 37 cases of PSCCB reported over a 6 year period (1985-2001) from one institution and compares their findings with cases of PSCCB entered into the Surveillance, Epidemiology and End Results (SEER) Database, which is a population based tumor registry sponsored by the National Cancer Institute and includes newly diagnosed cancer cases from 11 SEER participating areas. Out of 281,382 total numbers of breast cancer reported to the SEER database from 1988-2001, there were 137 cases of PSCCB reported [6]. A study by Nayak et al. identified 21 cases of PSCCB over a 15 year period (1985-2010) from one institution [9]. Additionally, Stevenson et al. identified seven cases of PSCCB out of 1647 cases of malignant breast tumors over a 48 year collection period (1945-1993), further demonstrating the rarity of this cancer [13].In addition, defining PSCCB can be challenging [13]. Currently, there are four pathological criteria that must be met prior to establishing the diagnosis of PSCCB according to Rosen’s Breast Pathology, which include: (1) tumor origin does not arise from overlying skin, nipple or adenexal components, (2) greater than 90% of the tumor must be composed of squamous cells, (3) no evidence of ductal or mesenchymal elements within the tissue sample, and (4) no other sites of primary squamous cell cancer are present [1-6,8-10,13,14]. However, there are questions as to whether PSCCB even exists or is simply a continuum with different degrees of squamous metaplasia within pre-existing adenocarcinoma [13]. This is exemplified in one study in which the ultrastructure of PSCCB is examined under light and electron microscopy and noted to have features of both glandular and squamous cells [13]. The World Health Organization (WHO) classifies PSCCB as a subgroup of epithelial-type metaplastic breast cancer, along with adenocarcinoma with spindle cell differentiation and adenosquamous carcinoma [9]. To further challenge things, the origin of PSCCB continues to remain unclear. In addition to Stevenson et al.’s hypothesis stating that PSCCB could be associated with pre-exisiting adenocaricionoma, there have been other hypotheses that the cancer arises from metaplasia of the ductal, luminal epithelial or myoepithelial cells, chronic abscesses or possible malignant transformation from epidermal or dermoid cyst [4,8,10].Detection of PSCCB can be very difficult as there is no specific physical exam or mammographic findings that are typical of this disease process. The majority of the patients present with a palpable mass. However, (name) presented a case report of a patient presenting with a pyogenic abscess. This patient presented with erythema and tenderness to her left breast and had an ultrasound study performed demonstrating a 7 x 6 x 2 cm cystic lesion. This lesion was subsequently drained, however, the patient returned 17 days later and underwent incision and drainage again with final pathology demonstrating PSCCB [3]. Unfortunately, because of its rarity, the diagnosis of PSCCB is usually made later in the disease process with the tumor being on average 4 cm, already putting them within the T2 category [1,10]. One author documented a wide range of tumor size within their study population, ranging from 1.2 to 13 cm’s [13]. In addition, on gross pathological exam, 50% of these lesions demonstrate a cystic component [1]. Axillary lymph node metastasis have been reported to range from 10-30% [2,7,10,12]. Hennessey et al.’s demonstrated a higher rate of lymph node metastasis at 50% in 33 patients with PSCCB [6]. PSCCB was noted to skip regional lymph node metastasis and present as a distant lesion at a rate of 30-33% and most notably to the lungs or soft tissue of the neck and mediastinum [2,10]. Furthermore, these tumors are most often estrogen receptor (ER), progesterone receptor (PR) and HER2/neu negative with overexpression of EGFR [7,10].In regards to imaging, there are no specific radiological findings of PSCCB that have been reported. A study by Sameuls et al. describes the mammography results of two patients with PSCCB as having an oval shape, indistinct, and spiculated mass without calcifications [15]. A second case report by Nair et al. describes their mammographic findings as having a rounded, dense mass with obscured margins and again, no calcifications [7]. These findings are similar to those of our cases that we presented. This further illustrates the elusiveness of this disease process. A fine needle aspiration and/or tissue biopsy needs to be performed in order to make a diagnosis of PSCCB [5]. On cytology, the principle finding is a predominant population of malignant squamous cells and may be enough data for the diagnosis. However, in rare cases of apocrine metaplasia, pseudosarcomatous stromal reaction, carcinosacroma and fibroadenoma with squamous cell metaplasia may cause a false positive and make diagnosis difficult [5,13].Most of the current literature portrays PSCCB as an aggressive cancer with a poor prognosis and outcome [6]. According to Behranwala et al., they found that larger tumor size (greater than five centimeters) and positive lymph node status were prognostic indicators of poor outcome. A study performed by Nayak et al. analyzed the histological features and disease outcome of 21 cases of PSCCB and noted that the predictors of decreased overall survival were patients over the age of 60 years old and >10% spindle cell tumor component [10]. They also demonstrated that lymph node status, mitotic rate, tumor necrosis, clear cell change and anaplastic features were not statistically significant in impacting overall survival or locoregional recurrence disease free survival [10]. The 5-year overall survival rate according to the SEER database on 137 patients with PSCCB from 1988-2001 was 64%. Hennessey et al. compared their institutions findings to the SEER database results and found similar overall survival rates with their median survival at 37 months and a 40% five year survival rate. They also reported that median relapse free survival with localized disease was 20 months [6].Since very little is known about the natural history of PSCCB, it often leads to multimodality therapy consisting of aggressive surgery followed chemotherapy. Initial surgical management of PSCCB is local disease control appropriate for tumor size with either total mastectomy or partial mastectomy with or without axillary clearance if the tumor to breast ratio is favorable [13]. Some investigators like Sheela et al. state that an axillary lymph node dissection should only be performed if there are clinically positive nodes or suspicious nodes present on imaging [12]. Other investigators including Behranwala et al. and Stevenson et al. report that while approximately 70% of patients with PSCCB do not have axillary lymph node metastasis on axillary dissection, complete axillary clearance should be routinely performed due to the unpredictability of the lymphatic spread [5,13].Chemotherapeutic choices remain a point of controversy as well as. Because the majority of these cancers are estrogen and progesterone receptor negative, there is no benefit to adjuvant hormonal therapy. In addition, PSCCB does not appear to respond adequately to conventional breast chemotherapy regimens. However, one report states that the use of CDDP, 5-FU and doxorubicin as a commonly used combination [13]. There have been reports that cisplatinum based chemotherapeutic agents have been successful in treating PSCCB [10,12,13]. The overexpression of EGFR has been used to the patient’s advantage as it has been reported to be more responsive to platinum and taxane chemotherapeutics [6,10,12,13]. With the use of anti-EFGR therapy in combination with taxane/platinum therapy needs to be further explored as it may make treating PSCCB more successful [10]. The use of neoadjuvent chemotherapy still remains unclear and according to Hennessey et al. has not demonstrated any proven benefit as 4 of their 33 patients received neoadjuvent therapy with no resultant tumor response [6]. However, according to a case report by Dejager et al, neoadjuvent cisplatinum and 5-FU was given to a patient with a 5.5 cm palpable PSCCB. After 3 cycles of chemotherapy, the mass was reported to be non-palpable and after surgical resection, the patient remained disease free for 2.5 years [4].The efficacy of radiation therapy in PSCCB is not well documented. Since locoregional relapse occurred frequently in the irradiated breast field, it is hypothesized that PSCCB is relatively radioresistant [6]. The lack of response to the radiation therapy may be due to the mixed cell type present within the cancer as proposed by Stevenson et al. [13]. In our case series, neither of our patients received systemic chemotherapy and has remained disease free at the time of their last follow up at five and eighteen month, respectively. Furthermore, one patient underwent radiation while the other patient did not, further exemplifying the need for more long term studies of this rare disease.In conclusion, due to the rarity of PSCCB, there is little consensus on a gold standard treatment or prognosis as evident by the different treatment methods discussed. Because treatment options can be limited, more research is needed to further elucidate the biological behaviors of this rare cancer as well as prognostic factors that may allow us to treat a patient more conservatively.