Karpas first described an endometrial cancer with psammomatous bodies in 1963 [
In 2009, our disease site team reviewed the literature and subsequently changed out treatment policy for women with Stage 1 UPSC. We moved from observation only to offering the sandwich technique of 3 cycles of carboplatin followed by brachytherapy and then 3 further cycles of carboplatin. Women with UPSC were identified by three strategies: the pathology database at Hamilton Health Sciences Centre (this includes all surgical cases done on site and all pathology consults from the local health integrated network), the cancer centre database (this includes all new patient referrals to the cancer centre), and a review of all weekly gynaecologic oncology tumour board meeting minutes from 2008 to present. The search terms included serous uterine cancer and UPSC. The charts of all cases were reviewed. FIGO staging was according to the 1988 classification. Only women who were Stage 1 by surgically staging with washings, hysterectomy, bilateral salpingoophorectomy, omentectomy, and retroperitoneal node dissection were included.
Disease characteristics included depth of invasion, amount of serous component and lymphovascular space involvement. Amount of serous component was defined as pure if the specimen contained at least 90% serous cancer and mixed if it did not.
Management included chemotherapy with platinum taxane usually with carboplatin AUC 5 based on 24 collection of urinary creatinine clearance and a 3-hour infusion of paclitaxel 175 mg/m2 q3 weeks. Radiation involved either vault brachytherapy to a dose of 1050 cGy and/or whole pelvis radiation to 4500 cGy.
Search strategy included MEDLINE (1980–2011), EMBASE (1980–2011) and the Cochrane Library databases for systematic reviews and clinical trials. Reference lists of papers and review articles were scanned for additional citations. Abstracts from 1997 to 2011 annual meetings of the American Society of Clinical Oncology (ASCO) were also searched. Canadian Medical Associate (CMA) Infobase (
We report on 26 Stage 1 UPSC cases who underwent surgery. Five are excluded as they did not have retroperitoneal node dissection. There were 6 women with Stage 1A disease. All patients had residual tumour in the uterus at the final pathology. There were 10 women with Stage 1B and 5 women with Stage 1C diseases.
In those with Stage 1A disease, only 1 patient received the recommended treatment, 2 only received chemotherapy and, 3 received no treatment. In those with Stage 1B disease, 5 received carboplatin and paclitaxel plus brachytherapy, 1 received carboplatin and paclitaxel, alone and 4 received no treatment. In those with Stage 1C disease, 3 received carboplatin with brachytherapy, 1 received carboplatin with pelvic radiation, 1 received carboplatin alone, and 1 had no treatment. There were 2/9 cases of treatment delay due to neutropenia in the group with the sandwich technique. There were 2/4 cases of treatment delay due to thrombocytopenia in the chemotherapy-only group. Our disease site team had a 50% compliance with the guideline.
All patients are alive at a median of 16-month followup (range 4–40, mean 16.4 mos). There were 2 recurrences at 12 and 13 months of followup. Both were vaginal recurrences in patients with Stage 1C disease. In both cases, the patients had not received upfront radiation therapy.
There was one prospective Phase 2 study addressing feasibility and toxicity of the sandwich technique of 3 cycles of chemotherapy then radiation and 3 further cycles of chemotherapy [
Most of the women in these studies began their management with surgical staging. Surgical staging involved hysterectomy, bilateral salpingoophorectomy, as well as retroperitoneal node dissection, omentectomy, and washings with or without peritoneal biopsies. The importance of surgical staging has been identified by several authors. Unlike endometrioid uterine cancer, in UPSC, metastatic disease cannot be reliably predicted based on uterine factors as women with apparent early disease (i.e., involving a polyp) may have already widespread intra-abdominal metastases. Thomas et al. [
In 12 studies, women received just adjuvant chemotherapy. In 13 studies, the women received some form of adjuvant radiation: (a) vault radiation, (b) pelvic radiation, (c) whole abdominal pelvic radiation (WART), or (d) a combination of radiation techniques. In 8 studies, women received a combination of chemotherapy and one or more radiation techniques.
There is a clear bias in the retrospective studies to provide women with deeper myometrial involvement with adjuvant therapy. For example, in Elit et al. [
Table
Characteristics of the Stage 1 UPSC studies included in this paper.
Author | Publication year | Single- or multi-institutional | Time period | Number of Stage 1 patients | Median FUP mos | |
---|---|---|---|---|---|---|
Bancher-Todesca et al. [ | 1998 | Multi | 1988–1995 | 3 | ||
Bristow et al. [ | 2001 | Single | 1989–1998 | 11 | ||
Carcangiu et al. [ | 1997 | Multi | 1987–1995 | 13 | ||
Dietrich et al. [ | 2005 | Multi | 1990–2003 | 29 | ||
Elit et al. [ | 2004 | Multi | 1985–2001 | 43 | ||
Fader et al. [ | 2009 | Multi | 1993–2006 | 142 | ||
Fields et al. [ | 2008 | Single | 1999–2004 | 16 | ||
Gallion et al. [ | 1989 | Single | 1973–1987 | 11 | ||
Gehrig [ | 2001 | Single | 1990–2000 | 6 | ||
Grice et al. [ | 1998 | Single | 1982–1993 | 12 | ||
Huh et al. [ | 2003 | Multi | 1987–2000 | 60 | ||
Havrilesky et al. [ | 2007 | Multi | 1976–2006 | 83 | ||
Kelly et al. [ | 2005 | Single | 1987–2004 | 74 | ||
Low et al. [ | 2005 | Single | 1994–2003 | 9 | ||
Piura et al. [ | 1998 | Multi | 1991–1997 | 7 | ||
Slomovitz et al. [ | 2003 | Single | 1989–2002 | 52 | ||
Thomas et al. [ | 2007 | Single | 1982–2005 | 42 | ||
Elit et al.* | 2011 | Single | 2008–2011 | 21 | ||
TOTAL |
*Data from Elit are not added to summary statistics as median followup is less than 24 months.
Disease characteristics of the patients in these studies.
Author | Stage (numbers) | Serous | Mixed | LVI (percent of cases) | |||||
1A | 1B | 1C | Total | + | − | Unknown | |||
Bancher-Todesca et al. [ | 1 | 1 | 1 | 3 | — | — | — | — | — |
Bristow et al. [ | 4 | 7 | — | 10 | — | — | — | — | — |
Carcangiu et al. [ | 13 | — | — | 13 | — | — | — | — | — |
Dietrich et al. [ | 7 | 17 | 5 | 29 | 16 | 5 | — | — | — |
Elit et al. [ | 22 | 13 | 8 | 43 | 36 | 7 | — | — | — |
Fader et al. [ | 51 | 65 | 26 | 142 | — | — | — | — | — |
Fields et al. [ | 2 | 12 | 2 | 16 | — | — | — | — | — |
Gallion et al. [ | 3 | 5 | 3 | 11 | — | — | 7 | 4 | — |
Gehrig [ | 6 | — | — | 6 | — | — | 5 | 1 | — |
Grice et al. [ | 4 | 4 | 4 | 12 | — | — | — | — | — |
Huh et al. [ | 19 | 26 | 15 | 60 | 51 | 10 | — | — | — |
Havrilesky et al. [ | 32 | 34 | 16 | 82 | — | — | 18 | 54 | 11 |
Kelly et al. [ | 33 | 29 | 12 | 74 | 47 | 27 | — | — | — |
Low et al. [ | 1 | 6 | 2 | 9 | — | — | — | — | — |
Piura et al. [ | 1 | 3 | 3 | 7 | — | — | — | — | — |
Slomovitz et al. [ | 19 | 26 | 7 | 52 | 27 | 23 | 12 | 40 | — |
Thomas et al. [ | 15 | 21 | 6 | 42 | — | — | — | — | — |
Elit et al.* | 6 | 10 | 5 | 21 | 19 | 2 | 3 | 18 | — |
TOTAL | 233 | 269 | 110 | 611 | 177/258 | 72/258 | 42/151 | 99/151 | 11/82 |
*Data from Elit are not added to summary statistics as median followup is less than 24 months.
Tables
Disease recurrences by disease characteristic.
Author | Stage | Serous | Mixed | LVI | ||||
1A | 1B | 1C | + | − | Unknown | |||
Bancher-Todesca et al. [ | 0/1 | 0/1 | 0/1 | — | — | — | — | — |
Bristow et al. [ | 1/ 4 | 1/7 | — | — | — | — | — | — |
Carcangiu et al. [ | 2/13 | — | — | — | — | — | — | — |
Dietrich et al. [ | 0/7 | 2/17 | 1/5 | 3/16 | 0/5 | — | — | — |
Fader et al. [ | 6/51 | 11/65 | 8/26 | — | — | — | — | — |
Fields et al. [ | 0 | 3/12 | 2/2 | — | — | — | — | — |
Gallion et al. [ | 0/3 | 4/5 | 1/3 | — | — | 5/7 | 0 | — |
Gehrig [ | 2/6 | — | — | — | — | — | — | — |
Grice et al. [ | 0/4 | 0/4 | 2/4 | — | — | — | — | — |
Havrilesky et al. [ | — | — | — | — | — | 18 | 53 | 11 |
Kelly et al. [ | 6/33 | 6/29 | 8/12 | — | — | — | — | — |
Low et al. [ | 0/1 | 1/6 | 0/2 | — | — | — | — | — |
Piura et al. [ | 0/1 | 0/3 | 1DOC/3 | — | — | — | — | — |
Slomovitz et al. [ | 4/19 | — | — | — | — | — | — | — |
Elit et al.* | 0/6 | 0/10 | 2/5 | 2/19 | 0/2 | 1/3 | 1/18 | |
21/143 (15%) | 31/149 (21%) | 22/56 (39%) | 19% | 0 | 22% | 65% | 13% |
*Data from Elit are not added to summary statistics as median followup is less than 24 months.
Description of management.
Author | Observe | RT | CT | CT | 3CT-RT-3CT | ||||
Vault | Pelvic | Pelvic + Brachy | WART | Other | Brachy | ||||
Bancher-Todesca et al. [ | 2 | — | — | 1 | — | — | — | — | — |
Bristow et al. [ | 10 | 1 | — | — | — | — | — | — | — |
Carcangiu et al. [ | 1 | — | — | — | 2 | — | — | 10 | — |
Dietrich et al. [ | — | — | — | — | — | — | 28 | — | — |
Elit et al. [ | 27 | — | 4 | — | 6 | — | 6 | — | — |
Fader et al. [ | 33 | — | — | — | — | 20 | — | 89 CT with or without RT | — |
Fields et al. [ | — | — | — | — | — | — | — | — | 16 |
Gallion et al. [ | 5 | — | — | 6 | — | — | — | — | — |
Gehrig [ | 6 | — | — | — | — | — | — | — | — |
Grice et al. [ | 5 | — | 4 | 2 | 1 | — | — | — | — |
Huh et al. [ | 40 | 4 | — | 5 | 3 | — | 7 | — | — |
Havrilesky et al. [ | 47 | 7 | 10 | 4 | 9 | — | 17 | — | — |
Kelly et al. [ | 11 | 3 | — | — | 1 | — | 21 | 8 | — |
Low et al. [ | — | — | — | — | — | — | — | 9 | — |
Piura et al. [ | 3 | — | — | 3 | — | — | 1 | 3 | — |
Slomovitz et al. [ | 27 | 6 | 2 | — | 3 | — | 5 | 1 | — |
Thomas et al. [ | 17 | 9 | 6 | — | 3 | — | 2 | 5 | — |
Elit et al.* | 8 | — | — | — | — | — | 4 | — | 9 |
234 | 30 | 26 | 21 | 28 | 20 | 87 | 125 | 16 |
*Data from Elit are not added to summary statistics as median followup is less than 24 months.
Recurrence rate by adjuvant treatment and stage of disease.
Author | Observation | Radiation | Chemotherapy | CT+RT | ||||||||
1A | 1B | 1C | 1A | 1B | 1C | 1A | 1B | 1C | 1A | 1B | 1C | |
Bancher-Todesca et al. [ | 0/1 | 0/1 | 0/1 | |||||||||
Bristow et al. [ | 1/ 4 | 1/1 | ||||||||||
Carcangiu et al. [ | 0/1 | 1/ 2 | 1/10 | |||||||||
Dietrich et al. [ | 0/7 | 2/17 | 1/5 | |||||||||
Elit et al. [ | 2/19 | 3/7 | 0/1 | 0/5 | 3/5 | |||||||
Fader et al. [ | 3/19 | 4/9 | 3/5 | 1/5 | 2/9 | 2/6 | 2/27 | 5/47 | 3/15 | |||
Fields et al. [ | 0/2 | 3/12 | 2/2 | |||||||||
Gallion et al. [ | 0/3 | 0/1 | 1/1 | 0/0 | 4/4 | 1/2 | ||||||
Gehrig [ | 2/6 | |||||||||||
Huh et al. [ | 1 /3 | 0/6 | 1 /3 | 0/2 | 0/3 | 0/2 | 0/1 | |||||
Low et al. [ | 0/1 | 1 /6 | 0/2 | |||||||||
Piura et al. [ | 1 /2 | 0/1 | 0/3 | 1/1 | 0/1 | 0/2 | ||||||
Thomas et al. [ | 0/11 | 0/3 | 0/1 | |||||||||
Elit et al.* | 0/3 | 0/4 | 1/1 | 0/2 | 0/1 | 1/1 | 0/1 | 0/5 | 0/3 | |||
Total | 8/64 | 8/19 | 4/9 | 3/13 | 1/29 | 7/16 | 0/10 | 2/20 | 2/8 | 3/42 | 9/65 | 5/21 |
Percent | 12.5% | 42% | 44% | 23% | 3% | 43.7% | 0 | 10% | 25% | 7% | 14% | 25% |
Stage 1A | 12.5% | 23% | 0 | 7% | ||||||||
Stage 1B | 42% | 3% | 10% | 14% | ||||||||
Stage 1C | 44% | 43.7% | 25% | 25% |
*Data from Elit are not added to summary statistics as median followup is less than 24 months.
Looking at the observation group alone, there is a substantial difference in recurrence rates between those with Stage 1A and the other Stages 1B-1C.
Looking at the individual studies, some authors have advocated
Looking at whether radiation improved survival over and above observation alone, the only group that appeared to benefit was the Stage 1B population. The individual studies addressed various forms of radiation therapy. Vault radiation has been used to decrease vaginal relapse. Pelvic radiation has been used to decrease locoregional recurrence. Whole abdominal pelvic radiation therapy has been used to sterilize the whole abdominal cavity (WART).
Huh et al. [
Looking at chemotherapy alone, again the numbers per strata are small, but there is a trend to markedly lower recurrences across the Stages 1A to 1C compared to observation alone or radiation. Several
The problem with chemotherapy alone was a concern for the high rate of recurrences especially at the vaginal apex. Looking at that chemotherapy and radiation data, the cell sizes are larger than the chemotherapy alone data. However, the rate of recurrences by substage look similar to the chemotherapy alone data and much better than the observation alone data. The individual studies on
Outcomes report by study.
Author | 5 yr OS | 5 yr PFS | Other | 1A | 1B | 1C |
---|---|---|---|---|---|---|
Bristow et al. [ | 82% | |||||
Carcangiu et al. [ | 83% | |||||
Dietrich et al. [ | 85.7% platinum + cyclo or platinum alone100% for carbo + taxol | |||||
Fields et al. [ | 3 yr OS 75% | |||||
Grice et al. [ | 84% | 48 mos′ | ||||
Huh et al. [ | 72% | 70% | ||||
Havrilesky et al. [ | Rads only | 3 yr OS 95%, PFS 85% | 3 yr OS 69%, PFS 59% | 3 yr OS 79%, PFS 60% | ||
Kelly et al. [ | 51 mos′ | 37 mos′ | 44 mos′ | |||
Low et al. [ | 73% | |||||
Piura et al. [ | 83.3% 3 yr for 10 patients only 7 surgically staged | 34 mos′ | 46 mos′ | 35 mos′ | ||
Slomovitz et al. [ | 63% | 5 yr OS 81.5% | 5 yr OS 58.6% | 5 yr OS 34.3% | ||
Thomas et al. [ | 85% | 78% | 5 yr OS 100% | 5 yr OS 89% | 5 yr OS 60% |
5 yr OS.
DFS: disease free survival.
′: Median survival.
A novel chemoradiation approach dubbed the “sandwich technique” was evaluated by Fields et al. [
We have pooled the data across various-case series in surgically staged women with Stage 1 UPSC. We show that the recurrence rates of those observed is significantly different in those with Stages 1B and 1C compared to those with Stage 1A disease. We show that recurrence rates are improved especially in those with Stage 1B and 1C diseases who receive adjuvant therapy including chemotherapy. To minimize vault recurrence, the addition of at least vault brachytherapy has been recommended. Our small case series reinforces this issue. However, the strata sizes in this study are too small to make this conclusion strongly. Although there is agreement to offer adjuvant treatment to women with Stage 1B and 1C diseases, the evidence is not as compelling as for the adjuvant treatment of those with Stage 1A disease. There appears to be a subgroup of Stage 1A patients without residual disease in the hysterectomy specimen which has a low chance of recurrence in whom observation seems appropriate.
There are significant limitations of this paper. We use the 1988 FIGO staging system in order to compare studies using the same staging classification. Given the new 1998 FIGO staging system, the old Stages 1A and 1B will be collapsed into Stage 1A. The old Stage 2A will now be absorbed within Stage 1. The subtleties of management and outcomes using the old staging system will be lost when categorizing patients with new system. The plethora of small case series exist because of the relative rarity of surgically staged 1 UPSC cases. Some of the issues that exist in trying to compare these studies include that many studies involved single versus multiple centres. Data collection in multicentre studies is more likely to be stringent as a data dictionary would have been used to define and collect similar data. Most of the studies are retrospective with the exception of prospective study by Fields. Thus, there is a strong risk of bias for treatment as we saw for the women with Stages 1B and 1C. In earlier reports, management results grouped UPSC with other high-grade cancers like clear cell; thus, it was difficult to discern the outcome of the UPSC cases and likely those cases were excluded. Many earlier studies reported on all stages of UPSC or a combination of Stages 1 and 2 versus focusing in on just Stage 1. Again, the information from these studies was likely excluded. The quality of the surgical staging varied across studies, that is, retroperitoneal node dissection only versus addition of omentectomy and peritoneal biopsies. Many earlier studies report their “experience with UPSC” and describe several adjuvant treatment strategies with small number of cases in each strata. Bias in choosing a management strategy is clearly an issue here. The radiation techniques varied across studies. Follow-up time in some studies was short, that is, 3 years; however, most recurrences were noted within the first 2 years after treatment, thus the impact of this short follow-up is not as concerning as may be the case in other disease types.
A randomized study provides the best opportunity for defining management. However, given the limitations listed about a randomized study in this rare entity is unlikely. This study of the current literature does provide themes around improvements for care in women with Stage 1 UPSC. There appears to be a role for chemotherapy in preventing the high rate of distant disease. There appears to be a role for some form of radiation, either vault brachytherapy or whole pelvic radiation to decrease the high risk of vault relapses.