Mary Atwood

Swedish Medical Center Seattle, Seattle, Washington, United States

Are you Mary Atwood?

Claim your profile

Publications (21)83.06 Total impact

  • J Malmgren · M Atwood · H Kaplan
    No preview · Article · Feb 2016 · Cancer Research
  • Henry G. Kaplan · Judith Malmgren · Mary Atwood
    No preview · Article · Aug 2013 · Cancer Research
  • Source
    Judith Malmgren · Mary Atwood · Henry Kaplan
    [Show abstract] [Hide abstract] ABSTRACT: Our research goal is to report on method of breast cancer detection among young women from a prospective cohort study of primary breast cancer patients, aged 20-49 years, 1990-2008 (n = 2579). Clinical presentation characteristics including race, TNM stage, first degree relative family history, histologic type and method of detection by patient (PtD), physician (PhysD), or mammography (MamD) were chart abstracted. Forward conditional stepwise regression was used to for association with detection method and Kaplan-Meier for relapse free survival (RFS) analysis. Among 20- to 39-year olds (n = 602) no change in detection method occurred over time with 12% MamD, 7% PhysD, and 81% PtD. Among 40- to 49-year olds, MamD BC increased over time (28% to 58%) and PtD BC decreased (63-36%) (Pearson X(2) = 72.72, p < .001). Among 20-39/40- to 49-year old MamD cases 31%/32% were stage 0 versus 2%/6% of the PhysD/PtD cases. In two separate conditional logistic regression models, older age at diagnosis and first degree relative BC history were associated with MamD BC for 20- to 39- and 40- to 49-year olds. Five-year MamD BC RFS was superior for both age groups (20-39: 94%, 40-49: 94%) compared to PtD BC rates (20-39: 80%, p = .016; 40-49: 88%, p < .001). For PtD BC 20- to 39-year olds had worse RFS (5 year 80%, 10 year 75%) than 40- to 49-year olds (5 year 88%, 10 year 82%) (p = .002) but RFS was equivalent for MamD cases by age. The majority of breast cancers among women 20-49 years were patient detected and mammography detection occurred rarely among youngest women. Lower stage and superior survival among MamD patients support mammography for detecting disease in high risk women aged 30-39 years and 40-49 years.
    Full-text · Article · Apr 2012 · The Breast Journal
  • [Show abstract] [Hide abstract] ABSTRACT: Cancer Program Practice Profile Reports (CP(3)R), established by the Commission on Cancer, are based on 6 guidelines for breast and colorectal cancer care using cancer registry data. The long-term goal is the use of cancer registry data for real-time interventions to optimize the process of individual patient multidisciplinary care. CP(3)R results using 593 analytic breast cancer cases in 2008 were compared in 3 databases: an institutional breast cancer research database, an institutional cancer registry, and a regional Cancer Surveillance System. Compliance with the CP(3)R guidelines calculated using the 3 databases was 80% to 98% for radiation therapy following breast-conserving surgery, 78% to 88% for combination chemotherapy of hormone receptor-negative stage T1c, II, or III disease, and 53% to 85% for hormone therapy of hormone receptor-positive stage T1c, II, or III disease. There was a high rate of discrepancy of tumor characteristics, treatment, and CP(3)R resulting from inaccurate and incomplete data. Using national cancer databases prospectively to monitor and ensure optimal multidisciplinary cancer care will require dramatic changes in cancer registry processes.
    No preview · Article · May 2011 · American journal of surgery
  • [Show abstract] [Hide abstract] ABSTRACT: Current guidelines recommend postmastectomy irradiation (PMI) for patients with tumors >5 cm and/or ≥4 positive lymph nodes. This study evaluates the effect of PMI on recurrence and survival within tumor size and node status groups. Locoregional and distant recurrences and survival for different tumor and treatment characteristics were analyzed in 2,797 patients with invasive breast cancer treated with mastectomy. Tumor size, positive nodes, extranodal extension, lymphatic/vascular invasion, estrogen receptor/progesterone negative, HER-2 positive, and high grade were associated with significantly increased recurrence. In patients with ≥4 positive nodes and patients with tumors >5 cm and positive nodes, PMI decreased local and distant recurrence (overall 53% vs 24%, P < .001) and increased disease-free survival (P < .001) but not overall survival. In patients with less disease, a benefit from PMI irradiation could not be identified. PMI is indicated for patients with ≥4 positive nodes or patients with tumors >5 cm and positive nodes.
    No preview · Article · May 2011 · American journal of surgery
  • Source
    [Show abstract] [Hide abstract] ABSTRACT: Human epidermal growth factor receptor 2 (HER2)/neu, topoisomerase II alpha (TOP2A), and polysomy 17 may predict tumor responsiveness to doxorubicin (DOX) therapy. We identified neoadjuvant DOX/cyclophosphamide treated breast cancer patients in our registry from 1997 to 2008 with sufficient tissue for testing (n = 34). Fluorescence in situ hybridization (FISH) testing was done on deparaffinized tissue sections pretreated using vendor's standard protocol modification, and incubated with US Food and Drug Administration approved Abbott Diagnostics Vysis PathVysion™ probe set, including Spectrum-Green-conjugated probe to α-satellite DNA located at the centromere of chromosome 17 (17p11.1-q11.1) and a Spectrum-Orange-conjugated probe to the TOP2A gene. Morphometric analysis was performed using a MetaSystems image analysis system. Manual counting was performed on all samples in which autofluorescence and/or artifact prevented the counting of sufficient numbers of cells. A ratio >2.0 was considered positive for TOP2A amplification. Polysomy 17 (PS17) presence was defined as signals of ≥2.5. Outcomes were pathological complete response (pCR), partial response (PR), and nonresponse (NR). Of 34 patients tested, one was TOP2A amplified (hormone receptor negative/HER2 negative, partial responder). The subset of TOP2A nonamplified, HER2 negative, and PS17 absent (n = 23) patients had treatment response: pCR = 2 (9%), PR = 14 (61%), and NR = 7 (30%). Including the two PS17 present and HER2-positive patients (n = 33), 76% of TOP2A nonamplified patients had pCR or PR. We observed substantial treatment response in patients lacking three postulated predictors that would be difficult to attribute to cyclophosphamide alone. Patients who are HER2 negative and lack TOP2A amplification and PS17 should not be excluded from receiving DOX-containing regimens.
    Full-text · Article · Aug 2010 · Cancer Management and Research
  • [Show abstract] [Hide abstract] ABSTRACT: Intraoperative identification of sentinel lymph node (SLN) metastases in breast cancer patients results in synchronous axillary lymph node dissection. We examined the effect of false-negative SLN biopsy on breast cancer treatments and recurrence rate. Patient and tumor characteristics, intraoperative and final SLN biopsy results, and treatments of patients with and without recurrence were compared. Recurrence rates for patients with true-positive SLN biopsy (9%) were significantly higher than rates for false-negative SLN biopsy patients (2%). Recurrence rates were significantly higher for patients with primary tumors greater than 2 cm, positive lymph nodes greater than 2 mm, and tumors with negative hormone receptors, and varied with treatment extent. Patients with greater amounts of disease in the breast and axilla required more treatment and had a higher recurrence rate. False-negative SLN evaluation occurred more commonly in patients with less lymph node metastasis and was not associated with an increased recurrence rate.
    No preview · Article · May 2010 · American journal of surgery
  • [Show abstract] [Hide abstract] ABSTRACT: This study reports the value of the tumor markers estrogen receptor (ER), progesterone receptor (PR), and human epidermal growth factor receptor 2 (HER2) in predicting the response of breast cancer to neoadjuvant chemotherapy. A community cancer center prospectively maintained breast cancer database containing over 8,000 patient records was used. Since 1989, 464 patients were treated with neoadjuvant chemotherapy followed by surgical resection and were tested for ER and PR. Estrogen receptor and/or PR positive patients were considered hormone receptor (HR) positive. Human epidermal growth factor receptor 2 status was available on 368 patients. Total, breast, and nodal pathologic complete response (pCR) rates, recurrence, and overall survival were assessed. Total and breast pCR rates were higher in HR negative (HR-) patients (26% and 32%, respectively) than in HR positive (HR+) patients (4% and 7%, respectively; p < 0.001). Compared to HR+ patients, HR- patients had higher recurrence rates (38% versus 22%; p < 0.001), a shorter time to recurrence (1.28 versus 2.14 years; p < 0.001), and decreased overall survival (67% versus 81%; p < 0.001). Human epidermal growth factor receptor 2 positive patients treated with neoadjuvant trastuzumab (NAT) demonstrated higher total pCR (34% versus 13%; p = 0.008), breast pCR (37% versus 17%; p = 0.02), and nodal pCR rates (47% versus 23%; p = 0.05) compared to HER2+ patients not treated with NAT. Furthermore, HER2+ patients who received NAT had lower recurrence rates (5% versus 42%; p < 0.001) and increased overall survival (97% versus 68%; p < 0.001). In conclusion, breast cancer HR status is predictive of total and breast pCR rates after neoadjuvant chemotherapy. Although HR- patients derive greater benefit from neoadjuvant chemotherapy in terms of pathologic response, they have worse outcomes in terms of recurrence and survival. Hormone receptor positive patients demonstrate significantly less response to neoadjuvant chemotherapy, but significantly better overall outcome. For both HR- and HR+, addition of NAT for HER2+ tumors results in both a superior response and outcome.
    No preview · Article · Apr 2010 · The Breast Journal
  • H. Kaplan · J. Malmgren · M. Atwood
    No preview · Article · Feb 2010 · Cancer Research
  • No preview · Article · Feb 2010 · Cancer Research
  • Source
    Henry G Kaplan · Judith A Malmgren · Mary Atwood
    [Show abstract] [Hide abstract] ABSTRACT: Adjuvant treatment of T1N0 breast cancer (BC) has evolved in recent years with chemotherapy options dependent on tumor size and cellular characteristics. Our goal is to describe the difference in outcome between T1N0 triple negative (TriNeg) and estrogen/progesterone receptor positive/her2/neu-negative BC. From our institute's registry, we identified primary BC patients diagnosed from 1998 to 2005, estrogen/progesterone receptor negative (ER-/PR-)/her-2/neu negative (her2-) (TriNeg = 110) and ER+/PR+/her2- (HR+/her2- = 919). Clinical diagnosis and treatment variables were chart abstracted. Vital and disease status were updated annually. Pearson chi-squared tests were used for bivariate analysis. Hazard ratios were calculated using the Cox proportional hazards model. Average patient age was 59 years, range 23-93 years and average length of follow-up was 4.22 years. T-stage distribution for HR+/her2- patients was 9% T1a (>0.1, < or = 0.5 cm), 34% T1b (>0.5 cm, < or = 1 cm), 57% T1c (>1 cm, < or = 2 cm) and for TriNeg, 6% T1a, 21% T1b, and 73% T1c. Sixty-five per cent of T1b and 73% T1c TriNeg patients received chemotherapy versus 7% of T1b and 32% of T1c HR+/her2- patients with TriNeg patients more likely to receive doxorubicin/cyclophosphamide/paclitaxel combined therapy. Recurrence rates were the following, T1b: 8.7%, TriNeg (2/23) versus 0%, HR+/her2- (0/315) and T1c: 8.8%, TriNeg (7/80) versus 2.1%, HR+/her2- (11/523). Five year relapse-free survival was 98% in the HR+/her2- group and 89% in the TriNeg group (log rank test = 27.77, p < 0.001). The hazard ratio for recurrence in the TriNeg group was 6.57 (95% CI = 2.34, 18.49) adjusted for age, tumor size, and adjuvant chemotherapy. Triple negative T1N0 patients have greater recurrence risk in spite of more aggressive therapy by both number treated and adjuvant chemotherapy type even in a low-risk category. New treatment modalities specific for triple negative disease are urgently needed.
    Full-text · Article · Sep 2009 · The Breast Journal
  • [Show abstract] [Hide abstract] ABSTRACT: In breast cancer treatment, sentinel lymph node (SLN) evaluation is used to identify patients who may benefit from axillary lymph node dissection (ALND). Intraoperative evaluation (IE) of SLNs facilitates immediate ALND. Controversy exists regarding the accuracy of intraoperative SLN evaluation for patients with invasive lobular carcinoma (ILC) compared to invasive ductal carcinoma (IDC). Using breast cancer registry data from January 2003 to March 2008, the intraoperative SLN evaluation of 66 ILC and 810 IDC patients was compared to the final SLN pathology result and to the performance of ALND. In ILC, the sensitivities of IE for isolated tumor cells (<or=.2 mm, N0[i+], n = 9), micrometastases (>.2 mm and <or= 2.0 mm, N1mi, n = 6), and macrometastases (>2.0 mm, N1a-3a, n = 21) were 0%, 17%, and 71%, respectively. The specificity was 100%. IE identified 16/27 (59%) of SLN-positive (N1mi, N1a-3a) axillae, resulting in synchronous ALND. Delayed ALND for false negative IEs (11/27, 41%) occurred in 7/11 patients (64%). In IDC, the sensitivities of IE for N0(i+) (n = 60), N1mi (n = 75), and N1a-3a (n = 129) metastases were 0%, 7%, and 71%, respectively. The specificity was 99.6%. IE identified 97/204 (48%) of SLN-positive (N1mi, N1a-3a) axillae, resulting in synchronous ALND. Delayed ALND for false negative IEs (107/204, 52%) occurred in 38/107 patients (36%). Sensitivity and specificity of intraoperative SLN evaluation is very similar in ILC and IDC patients. Intraoperative SLN evaluation facilitated synchronous ALND in concordance with recommended practice guidelines.
    No preview · Article · Jun 2009 · American journal of surgery
  • [Show abstract] [Hide abstract] ABSTRACT: This study examines the downstaging of breast cancer axillary lymph node (ALN) metastasis by neoadjuvant chemotherapy (NCT) and the potential facilitation of axillary-conserving surgery. Records of breast cancer patients treated with NCT, breast surgery, and pathological ALN assessment were reviewed using the institutional breast cancer database. Of 473 cases, 309 (65%) were clinically ALN-positive (cN+) and 164 (35%) were clinically ALN-negative (cN-). Pre-NCT, needle biopsy, sentinel lymph node (SLN) biopsy, or axillary dissection documented metastasis in 94% (117/124) of cN+ and 27% (13/49) of cN- patients tested. Pathological complete response of ALNs to NCT was documented in 36% (41/115) of patients. False negative SLN biopsy following NCT occurred in 4% of cases (1/28). NCT downstages primary breast cancer and ALN metastasis. ALN and SLN biopsy following, rather than before, NCT facilitate both breast- and axillary-conserving surgery.
    No preview · Article · Apr 2009 · American journal of surgery
  • [Show abstract] [Hide abstract] ABSTRACT: Background In 2003, the American Joint Committee on Cancer (AJCC) initiated the 6th edition staging criteria, including pN0(i+) and pN1mi categories for breast cancer. However, the clinical significance of these categories is debated in the literature. Methods A prospective registry was used to identify patients staged with sentinel lymph node (SLN) biopsy. SLN evaluation included routine serial sectioning and immunohistochemical stains. SLN biopsies performed before January 2003 were restaged according to the AJCC’s 6th edition criteria. Results Of 954 SLN biopsies identified, on review, 491 N0i-, 86 N0i+, 73 N1mi, 146 N1a, 29 N2a, and 11 N3a patients were available for analysis with a median follow-up of 45.4 months. Significant prognostic and therapeutic differences existed between the groups. Differences in overall survival (OS) and recurrence-free survival (RFS) were only noted when the size of the metastases reached the N1a level. There were no statistically significant differences in OS or RFS between N0(i−) and N0(i+) or N1mi disease. Cases that were N0(i+) or N1mi were more likely to have other poor prognostic factors and to receive more aggressive therapy. Conclusion SLN biopsy allows a more sensitive evaluation of lymph nodes for metastatic cells. This has led to the increased identification of very small axillary metastases. While the new microstaging categories are not yet clearly associated with a significantly decreased OS or DFS in this series, they are associated with other poor prognostic factors and more local/regional and systemic therapy. Further analysis of the microstaging categories is needed.
    No preview · Article · Dec 2008 · Annals of Surgical Oncology
  • [Show abstract] [Hide abstract] ABSTRACT: Optimal breast cancer control outcomes include early diagnosis, thorough staging work-up, and lowest disease-related mortality. The standardized case-mix index (SCI) for breast cancer was derived from the stage distribution of cases weighted for earlier diagnosis using the national stage 5-year mortality provided in the National Cancer Data Base expressed as an institution or region (observed) to national (expected) ratio (O/E). The standardized work-up index was the mean O/E mortality ratio for each stage. The standardized treatment index was the total O/E 5-year mortality corrected for the SCI. The overall performance evaluation (OPE) was SCI x standardized work-up index x standardized treatment index. Institutional data were acquired from a prospectively maintained breast cancer database. OPE scores for 42 states acquired from the National Cancer Data Base tended to be best for the northeast and northwest states. Washington and Oregon OPE scores were in the top 20%. The Swedish Cancer Institute OPE score improved steadily from 1990 to 2000 and was better than the Washington state score. By calibrating breast cancer stage of diagnosis and mortality to a national standard, regional and institutional breast cancer control programs can be compared objectively.
    No preview · Article · Jun 2008 · American journal of surgery
  • [Show abstract] [Hide abstract] ABSTRACT: In breast cancer treatment, intraoperative sentinel lymph node (SLN) evaluation is used to identify patients who may potentially benefit from immediate completion of axillary lymph node dissection. Prospectively collected breast cancer registry data identified 516 SLN biopsies between January 2003 and December 2005. Intraoperative evaluation (IE) of the SLNs was performed in 479 axillae. Final pathology by hematoxylin and eosin and, for negative nodes, by immunohistochemical stains was compared with the IE result. The effect of IE and final pathology on surgical treatment was examined. The sensitivities for IE of N0(i+) (n = 39), N1mi (n = 41), and N1a-3a (n = 89) metastases were 0%, 5%, and 63%, respectively. The specificity was 99.7%. IE identified 57 (44%) of SLN-positive (N1mi and N1a-3a) axillae, thus resulting in synchronous axillary lymph node dissection for those patients. Reoperation for false-negative IEs (N1mi or N1a-3a with negative IE) occurred in only 27 axillae (39%). IE of SLNs has adequate sensitivity and excellent specificity. In addition to allowing patients to benefit from synchronous surgery, IE helped patients to receive care in concordance with recommended practice guidelines. The false-negative IE of SLNs highlights uncertainty with the clinical significance of axillary nodal staging when only small amounts of metastatic disease are identified in the axilla.
    No preview · Article · Apr 2007 · Annals of Surgical Oncology
  • [Show abstract] [Hide abstract] ABSTRACT: Metaplastic breast carcinoma (MBC) is a rare poorly differentiated breast cancer characterized by coexistence of ductal carcinoma with areas of matrix producing, spindle-cell, sarcomatous, or squamous differentiation; ER/PR/HER2 negativity; and a reputation for poor outcome. The Swedish Cancer Institute prospective breast cancer database (> 6500 patients; 1990-2005) has 24 MBC cases that were compared with typical breast cancer cases matched for age, date of diagnosis, stage, and ER/PR/HER2 status. The mean metaplastic primary tumor diameter was 2.5 cm. The histological/nuclear grade was high in 21 of 24 cases. No patient had distant metastasis. ER and/or PR receptor status was negative in all cases. HER2 was negative in 10 of 11 cases tested. EGFR (HER1) was positive in 7 of 7 cases tested. All patients had sentinel and/or axillary lymph node dissection and surgical resection; 18 received chemotherapy and 22 had radiation therapy. Four patients had distant recurrences 5 to 88 months from diagnosis. Five-year survival was 83% (95% confidence interval, 66-100%). Comparison with matched typical breast cancer cases revealed no major significant difference in multidisciplinary treatment patterns, recurrence, or survival. MBC is associated with poor prognostic indicators, but outcomes comparable with matched typical breast cancer cases can be achieved with routine aggressive multidisciplinary care. Increased, expression of EGFR (HER1) provides an opportunity for targeted tumor therapy.
    No preview · Article · Jun 2006 · The American Journal of Surgery
  • Source
    Henry G Kaplan · Judith A Malmgren · Mary Atwood
    [Show abstract] [Hide abstract] ABSTRACT: The results of randomized clinical trials have suggested that after receiving radiotherapy and/or chemotherapy, patients with primary breast carcinoma have an increased risk of developing leukemia. In the current study, the authors set out to assess the reported association between breast carcinoma treatment and leukemia risk. A registry of all patients with breast carcinoma who were treated at a community-based institution since 1989 (updated annually for recurrence and/or vital status) was linked to the National Cancer Institute Surveillance, Epidemiology, and End Results database to confirm complete ascertainment of leukemia cases occurring within this registry population. Incidence rates were calculated for women who were treated for primary Stage 0-III breast carcinoma and had a follow-up duration of > or = 24 months (n = 2866). Patients who did not undergo surgery (n = 5), patients for whom chemotherapy records were incomplete or who received nonstandard chemotherapy regimens (n = 69), patients who underwent stem cell transplantation (n = 83), and patients who were lost to follow-up or who had unknown disease status at follow-up (n = 81) were excluded from the analysis (total, n = 238). Among patients diagnosed with breast carcinoma between 1992 and 1999, the crude overall leukemia incidence rate was 0.28%, and the acute myelogenous leukemia (AML)/myelodysplastic syndrome (MDS) incidence rate was 0.11%. The average follow-up duration was 5.46 years (minimum, 2 years). Eight incident cases of leukemia were documented (2 cases of AML, 1 case of acute lymphoblastic leukemia, 1 case of MDS/refractory anemia with excess blasts, 2 cases of chronic myelogenous leukemia, and 2 cases of chronic lymphocytic leukemia). National age-adjusted overall leukemia incidence rates for the period 1996-1998 predict the occurrence of 9 cases (incidence rate, 0.31%) in the current cohort of women ages 21-94 years. The incidence of leukemia by treatment category was as follows: no surgery/no chemotherapy/no radiotherapy, 2 of 154 patients (1.30%); surgery/no chemotherapy/radiotherapy, 4 of 1403 patients (0.29%); surgery/chemotherapy/no radiotherapy, 0 of 352 patients (0%); and surgery/chemotherapy/radiotherapy, 2 of 957 patients (0.21%). In contrast to findings reported from previous randomized clinical trials, the authors did not find evidence of increased posttreatment leukemia incidence in association with the use of chemotherapy, including doxorubicin-based regimens.
    Full-text · Article · Oct 2004 · Cancer
  • No preview · Article · Dec 2003 · Gynecologic Oncology
  • [Show abstract] [Hide abstract] ABSTRACT: To determine if the type of operative incision influences the adequacy of surgical staging in patients with uterine cancer. All patients with uterine cancer referred to the Swedish Medical Center Cancer Institute for adjuvant radiotherapy between June 1, 1989, and June 1, 1999, who underwent comprehensive surgical staging and for whom complete records could be obtained were eligible. Data on type of incision, weight, medical comorbidities, histology, total number and distribution of lymph nodes (LN), estimated blood loss, complications, and length of stay were abstracted retrospectively. Statistical analysis with two-tailed Student t test, chi(2), Fisher's exact, and Kaplan-Meier survival curves were performed. Five hundred four women with uterine cancers were referred to the Cancer Institute with 332 meeting inclusion criteria. A vertical midline incision (ML) was used in 236 (72%) while 96 (28%) received a Pfannenstiel incision (PI). No panniculectomies were performed. There were no statistically significant differences in age, weight, stage, histology, comorbidities, or estimated blood loss between the ML and PI groups. ML was associated with significantly more intraoperative and postoperative complications (34 vs. 7; P = 0.003). When compared to ML a greater number of total LN (21.0 vs. 16.8; P = 0.001) and a comparable number of pelvic LN (13.7 vs. 12.2; P = 0.14) were procured through a PI. More patients with a ML (72% vs. 63%; P = 0.13) had para-aortic lymph nodes (PALN) dissected; however, when obtained equivalent numbers of nodes were removed (3.52 vs. 4.36; P = 0.14). Overall, the median length of stay was statistically shorter for those patients operated on via a PI (4 vs. 3 days; P = 0.007). The projected 5-year disease-free (83% vs. 85%) and disease-specific (87% vs. 85%) survival was unaffected by incision. In the heaviest quartile of patients (>180 lb), a statistically greater number of total LN (23.3 vs. 16.5; P = 0.005) and pelvic LN (16.7 vs. 11.5; P = 0.05) were obtained with a PI. Again, PALN were sampled more frequently (67% vs. 56%; P = 0.45) in patients with a ML, but the mean LN yield was no different (3.91 vs. 5.20; P = 0.37). Likewise, in this heaviest quartile, there were no statically significant differences in operative complications (7 vs. 1; P = 0.43) with either incision. Comprehensive surgical staging for uterine cancers can be adequately performed through a PI without greater morbidity or mortality. By using this surgical approach, patients with uterine cancer can benefit from the inherent benefits previously described for PI. Appropriate patient selection, however, is necessary.
    No preview · Article · Mar 2003 · Gynecologic Oncology