All IPC cases were analyzed with relation to patient gender and age at presentation as well as tumor stage. The CCR classifies IPC as either in-situ or invasive as determined by the local pathologist. The invasive tumors are then further classified as localized, regional, or metastatic. The Berkson-Gage life table method was used to calculate observed cumulative survival and relative cumulative survival rates. Relative survival rates compare the mortality of a group of patients to a similar group of people from the general population. Relative survival measures the total impact of the disease on survival, regardless of what was considered to be the cause of death. The survival rates of the two histologically-derived subgroups of IPC were compared to each other.

As a comparison, the observed cumulative survival rate and relative cumulative survival rate of all other (non-IPC) invasive breast cancers in the CCR were calculated for the same time period.

Nominal data were compared using a chi-square and survival data were compared using a Z-test. A p-value of <0.05 was considered statistically significant.

Forty-seven percent of cases of IPC were classified as carcinoma in-situ (CIS) (n=427), while 53% of cases were classified as invasive (n=490). Of these invasive cases, the vast majority was localized at the time of diagnosis (89.6%, n=439), while 7.8% of the cases were classified as regional disease, with either direct extension into adjacent tissue or axillary lymph node involvement (n=39). Only 2 cases were reported as distant stage at the time of diagnosis (0.4%).

The overall observed cumulative survival of patients with IPC was 82.0% at 5 years and 61.2% after 10 years. When the sub-types were divided, there was a slightly significantly better observed cumulative survival of patients with in-situ IPC at 5 years compared to those with invasive IPC (85.0% vs 75.0%, P=0.05). At 10 years, however, patients with CIS and invasive IPC had a virtually identical cumulative survival (61.7% and 60.6%, p=0.08) (Figure 1).

In order to look specifically at the effect of this breast malignancy on survival, we calculated relative cumulative survival rates, which compare the survival of patients with the breast tumor to age-matched controls from the general population. The relative survival of all patients with IPC, both CIS and invasive, was 97.3% after 5 years and 95.6% after 10 years. There were no significant differences in the relative survival rates between patients with invasive IPC compared to those with in-situ IPC at either 5 or 10 years. At 5 years, the relative cumulative survival of patients with in-situ IPC and invasive IPC were 101.3% and 93.9%, respectively (p=0.18). After 10 years, patients with in-situ IPC had a relative cumulative survival of 96.8%, while those with invasive IPC had a relative cumulative survival of 94.4% (p=0.75) (Figure 2).

IPC is generally perceived as a malignancy of elderly women.⁹ Indeed, in this study, though IPC affected patients of all ages, most patients were older, with the median age being 69.5 years old. On the other hand, though the majority of the patients with the tumor were female, there were 32 cases (3.5%) diagnosed in men. This rate is slightly higher than the rate of invasive ductal carcinoma and DCIS seen in the male population and higher than previously described in the literature.^{10, 11} The majority of the larger published case series analyzing IPC do not include any male patients; however the cases in the CCR show that IPC is not an entity unique to women and that breast cysts in men may occasionally contain a malignant component.^{1, 2, 12, 13}

The histological classification of IPC is confusing, and there is controversy within the surgical and pathology literature regarding its categorization and nomenclature.^{5, 7, 8} In 1980, a classification system was described by Fisher et al. that divided papillary carcinomas into the invasive and non-invasive forms.³ The non-invasive form was then further subdivided into the diffuse form, the papillary variant of DCIS, involving multiple small and medium sized ducts, and the localized form, termed “intracystic” or “encysted” papillary carcinoma.^{3, 12} This localized form, IPC, describes a solitary tumor in an encysted or dilated duct.³ The presence of a layer of myoepithelial cells (MECs) at the periphery of areas of papillary carcinoma have historically been used to define a lesion as in-situ rather than invasive.⁸ Recent studies have shown that, in contrast to papillary DCIS, IPC does not appear to have an MEC layer surrounding the tumor nodules.^{5, 7, 8} This observation has lead to the idea that, in some cases, IPC may not actually be an in-situ carcinoma but may actually be an encapsulated nodule of low-grade invasive carcinoma or may be part of a spectrum of progression intermediate between in situ and invasive disease.^{5, 7, 8} Some pathologists now prefer the term “encapsulated papillary carcinoma” to the more traditional “intracystic papillary carcinoma.”^{7, 8, 14} Within the CCR, IPC lesions have been reported by the local pathologist as either in-situ or invasive. Within this study, the tumors were determined to be in-situ in approximately half the cases. Local pathologists may report the tumor as invasive based on the lack of an MEC layer or the presence of entrapped neoplastic cells within the fibrous capsule, which often have the appearance of invasive carcinoma.⁸ Without access to the original specimen, it was not possible to determine the criteria used to establish the diagnosis of an invasive lesion. While of academic interest, this distinction is likely to be of little clinical relevance. ^{7, 8, 15}

IPC has been perceived to have a good prognosis.^{2, 12, 15} The CCR illustrates an observed cumulative survival of over 60% at 10 years. Since most patients with IPC are older, this overall survival rate may be influenced by their older age and consequent co-morbidities. Therefore, in order to look specifically at the impact of IPC on survival, we calculated relative cumulative survival rates. Relative survival is a useful way to examine the effect of a particular disease process by comparing patients with the disease to age-matched controls from the general population. Because it is obtained by adjusting observed survival for the normal life expectancy of the general population of the same age, the relative survival rate is an estimate of the chance of surviving the effects of cancer. This is particularly useful for cancers that effect an older patient population, such as IPC. The CCR data demonstrate that, at 5 years, patients who presented with IPC without evidence of invasion had a survival rate that exceeded that of the general population. At 10 years, this survival rate remained greater than 95%. Even those patients who were diagnosed with invasion at the time of diagnosis had a relative survival rate of over 90% of that of the general population at both 5 and 10 years. These data support previous studies that have shown a low mortality for patients with IPC. Several case series have identified no cancer-related deaths from IPC among patients followed for over 50 months.^{2, 12, 15} Lefkowitz et al. reviewed the charts of 29 patients diagnosed with IPC and found only one patient was suspected to have died from her cancer. Her tumor was associated with an invasive ductal carcinoma.¹³ In their analysis of 45 patients with IPC, Fayanju et al. observed only one patient who had a cancer-related death. This patient had an associated DCIS.¹ The CCR data corroborates these studies while also demonstrating that there is a good outcome from IPC whether the tumor is deemed in-situ or invasive. This excellent long-term prognosis for IPC contrasts that of other non-IPC breast cancers, the majority of which are invasive ductal carcinoma. The CCR data over the same 18 year time period demonstrates both a lower observed and lower relative survival rate for patients with non-IPC breast cancers. Clearly, patients with IPC are much less likely to die than those diagnosed with the more common types of breast cancer, such as invasive ductal carcinoma.

There are no clear guidelines on how to approach the management of IPC and there are varying trends over the last 25 years as to how to treat the tumor.² It was not possible to analyze outcomes as they relate to treatment because the CCR does not contain inclusive information regarding treatment for all years. Several studies, however, have shown that prognosis is excellent and recurrence rates low, regardless of the intervention.^{1, 2, 15} The mainstay of treatment is surgical excision. Breast-conserving surgery and mastectomy (simple, modified radical, or radical) have been utilized in the treatment of IPC. The majority of studies fail to show a difference in survival depending on which surgical procedure is performed, with a good prognosis with all interventions.^{1, 2, 12} The surgical management of IPC seems to parallel that of invasive ductal carcinoma, with a lower rate of breast conservation in older studies. For instance, in a study by Carter et al. which analyzed 41 cases of IPC treated prior to 1974, 70% of patients with a diagnosis of IPC from excisional biopsy underwent mastectomy, with over a third of these patients having a radical mastectomy.¹² None of these patients had a recurrence or cancer-related death. The remaining 12 patients did not have any further surgical intervention. There were 3 tumor recurrences in this group, but no cancer-related deaths, and all of these recurrences occurred in patients with concurrent DCIS. More recent studies show a tendency towards breast conservation with no significant increase in tumor recurrence or cancer-related deaths.^{1, 2, 15}

In addition to surgical excision, several studies have reported on the use of adjuvant radiation and/or endocrine therapy in the management of IPC.^{1, 2, 15} In their retrospective review of the treatment and outcome of 40 patients with IPC, Solorzano et al. observed that 30% of the patients had received adjuvant radiotherapy. The addition of radiation to the treatment of patients did not change the incidence of local recurrence or likelihood of death compared to those who did not receive radiation.² Within this study, no criteria for how patients were chosen to receive radiation were evident. Additionally, there has been no clear indication for adjuvant endocrine therapy, even among patients with estrogen-receptor positive tumors. Furthermore, the addition of hormonal treatment does not appear to have impacted outcome.^{1, 15} Given the overall good prognosis and that no adjuvant treatment has been shown to increase cancer-free survival, there is concern for a potential overtreatment in the disease.^{5, 15}

There is no consensus as to the role of axillary staging procedures for IPC. Within the CCR, less than 8% of IPC cases were reported to have regional disease, though as it is not reported what percentage of patients underwent axillary sampling, this rate may not be accurate. Most previous studies show a similarly low rate of lymph node involvement, though axillary dissections were not consistently performed in any study and there do not appear to be any concrete or universal indications for axillary dissection.^{2, 12, 13, 15} The CCR data, as well as several other studies, however, do demonstrate occasional axillary lymph node involvement and other authors have reported axillary lymph node involvement even in cases without reported adjacent invasive ductal carcinoma.^{2, 16}

Given the controversial and evolving understanding of the invasiveness of the tumor and the occasional axillary lymph node metastasis associated with IPC, we recommend treating IPC like DCIS with microinvasion. While, many authors have suggested that IPC, whether labeled invasive or CIS, be treated like DCIS,^{5, 15} it has been reported that, like IPC, DCIS with microinvasion has between an 8–14% incidence of axillary lymph node involvement.^17–19 Sentinel lymph node biopsy has, therefore, been advocated for these patients. ^17–19 Sentinel lymph node biopsy has not been assessed in IPC, but it seems to be the practical and prudent way to evaluate the axilla in patients with IPC.^{2, 16}

There are several limitations inherent in this study. First, it relies on the accurate reporting of local pathologists. Second, it is well documented that IPC may be associated with adjacent areas of DCIS and invasive ductal carcinoma. There are no data within the CCR as to which cases were isolated IPC and which had an associated lesion, although if there was frank invasive ductal carcinoma, the cases would have been recorded as invasive ductal carcinoma and not as IPC. Data available from other series suggest, however, that though associated pathology may help guide treatment, its presence has little effect on overall survival.^{2, 12} Finally, given the great controversy within the literature about the histologic classification of IPC, it is unclear what methods the pathologists are using to determine the histology of the tumors, specifically whether the tumor was in-situ or invasive. The CCR results demonstrate, however, that this distinction is likely not of clinical significance because, regardless of the classification, the prognosis is excellent.