Cost-effectiveness of prenatal screening methods for congenital heart defects in pregnancies conceived by in-vitro fertilization.

OBJECTIVES: To determine if a policy of universal fetal echocardiography (echo) in pregnancies conceived by in-vitro fertilization (IVF) is cost-effective as a screening strategy for congenital heart defects (CHDs) and to examine the cost-effectiveness of various other CHD screening strategies in IVF pregnancies. METHODS: A decision-analysis model was designed from a societal perspective with respect to the obstetric patient, to compare the cost-effectiveness of three screening strategies: (1) anatomic ultrasound (US): selective fetal echo following abnormal cardiac findings on detailed anatomic survey; (2) intracytoplasmic sperm injection (ICSI) only: fetal echo for all pregnancies following IVF with ICSI; (3) all IVF: fetal echo for all IVF pregnancies. The model initiated at conception and had a time horizon of 1 year post-delivery. The sensitivities and specificities for each strategy, the probabilities of major and minor CHDs and all other clinical estimates were derived from the literature. Costs, including imaging, consults, surgeries and caregiver productivity losses, were derived from the literature and Medicare databases, and are expressed in USA dollars ($). Effectiveness was quantified as quality-adjusted life years (QALYs), based on how the strategies would affect the quality of life of the obstetric patient. Secondary effectiveness was quantified as number of cases of CHD and, specifically, cases of major CHD, detected. RESULTS: The average base-case cost of each strategy was as follows: anatomic US, $8119; ICSI only, $8408; and all IVF, $8560. The effectiveness of each strategy was as follows: anatomic US, 1.74487 QALYs; ICSI only, 1.74497 QALYs; and all IVF, 1.74499 QALYs. The ICSI-only strategy had an incremental cost-effectiveness ratio (ICER) of $2 840 494 per additional QALY gained when compared to the anatomic-US strategy, and the all-IVF strategy had an ICER of $5 692 457 per additional QALY when compared with the ICSI-only strategy. Both ICERs exceeded considerably the standard willingness-to-pay threshold of $50 000-$100 000 per QALY. In a secondary analysis, the ICSI-only strategy had an ICER of $527 562 per additional case of major CHD detected when compared to the anatomic-US strategy. All IVF had an ICER of $790 510 per case of major CHD detected when compared with ICSI only. It was determined that it would cost society five times more to detect one additional major CHD through intensive screening of all IVF pregnancies than it would cost to pay for the neonate's first year of care. CONCLUSION: The most cost-effective method of screening for CHDs in pregnancies following IVF, either with or without ICSI, is to perform a fetal echo only when abnormal cardiac findings are noted on the detailed anatomy scan. Performing routine fetal echo for all IVF pregnancies is not cost-effective. © 2020 International Society of Ultrasound in Obstetrics and Gynecology.

Quality of life is significantly associated with survival in women with advanced epithelial ovarian cancer: An ancillary data analysis of the NRG Oncology/Gynecologic Oncology Group (GOG-0218) study.

OBJECTIVE: Evaluate association between baseline quality of life (QOL) and changes in QOL measured by FACT-O TOI with progression-free disease (PFS) and overall survival (OS) in advanced epithelial ovarian cancer (EOC). METHODS: Patients enrolled in GOG-0218 with completed FACT-O TOI assessments at baseline and at least one follow-up assessment were eligible. Baseline FACT-O TOI scores were sorted by quartiles (Q1-4) and outcomes compared between Q1 and Q2-4 with log-rank statistic and multivariate Cox regression adjusting for age, stage, post-surgical residual disease size, and performance status (PS). Trends in FACT-O TOI scores from baseline to the latest follow-up assessment were evaluated for impact on intragroup (Q1 or Q2-4) outcome by log-rank analysis. RESULTS: Of 1152 eligible patients, 283 formed Q1 and 869 formed Q2-4. Mean baseline FACT-O TOI scores were 47.5 for Q1 vs. 74.7 for Q2-4 (P<0.001). Q1 compared to Q2-4 had worse median OS (37.5 vs. 45.6months, P=0.001) and worse median PFS (12.5 vs. 13.1months, P=0.096). Q2-4 patients had decreased risks of disease progression (HR 0.974, 95% CI 0.953-0.995, P=0.018), and death (HR 0.963, 95% CI 0.939-0.987, P=0.003) for each five-point increase in baseline FACT-O TOI. Improving versus worsening trends in FACT-O TOI scores were associated with longer median PFS (Q1: 12.7 vs. 8.6months, P=0.001; Q2-4: 16.7 vs. 11.1months, P<0.001) and median OS (Q1: 40.8 vs. 16months, P<0.001; Q2-4: 54.4 vs. 33.6months, P<0.001). CONCLUSIONS: Baseline FACT-O TOI scores were independently prognostic of PFS and OS while improving compared to worsening QOL was associated with significantly better PFS and OS in women with EOC.

A personalized paradigm in the treatment of platinum-resistant ovarian cancer - A cost utility analysis of genomic-based versus cytotoxic therapy.

OBJECTIVE: To assess the cost-effectiveness of a strategy employing genomic-based tumor testing to guide therapy for platinum-resistant ovarian cancer. METHODS: A decision model was created to compare standard of care (SOC) cytotoxic chemotherapy to a genomic-based treatment strategy. The genomic arm included tumor testing with treatment directed at targets identified. Overall survival was assumed to be similar between strategies; quality of life (QOL) was assumed superior during targeted therapy compared to chemotherapy. Pertinent uncertainties (cost of targeted therapy and genomic testing, response to targeted therapy, probability of a tumor having a targetable alteration, and impact on QOL) were evaluated in a series of one-and two-way sensitivity analyses. RESULTS: The genomic testing strategy was more expensive ($90,271 vs. $74,926) per patient than SOC. The incremental cost-effectiveness ratio (ICER) of the genomic strategy was $479,303 per quality-adjusted life year saved (QALY). Model results were insensitive to the cost of genomic testing, differences in QOL, and the probability of identifying a targetable alteration. However, the model was sensitive to the cost of targeted therapy. For example, when the cost of targeted therapy was reduced to 56% of its current cost (or $6400/cycle), the genomic strategy became more cost-effective with an ICER of $96,612/QALY. CONCLUSIONS: Genomic-based tumor testing and targeted therapy in patients with platinum-resistant ovarian cancer is not cost-effective compared with SOC. However, reducing the cost of targeted therapy (independently, or in combination with reducing the cost of the genomic test) provides opportunities for improved value in cancer care.

Selective cardiac surveillance in patients with gynecologic cancer undergoing treatment with pegylated liposomal doxorubicin (PLD).

OBJECTIVE: The study objective was to examine the safety and cost savings of selective cardiac surveillance (CS) during treatment with pegylated liposomal doxorubicin (PLD). METHODS: A retrospective, dual institution study of women receiving PLD for the treatment of a gynecologic malignancy was performed. The study period was 2002-2014. At both institutions, a selective strategy for CS was implemented in which only high-risk women with a cardiac history or with symptoms suggestive of cardiac toxicity during PLD treatment underwent a cardiac evaluation. Patient demographics, clinical and treatment history were evaluated. Cost analyses were performed utilizing professional/technical fee rates for echocardiogram and multi-gated acquisition scan for each state. RESULTS: PLD was administered in 184 women. The mean patient age was 62.7years, and 79% were treated for recurrent ovarian or peritoneal carcinoma. The median cumulative administered dose of PLD was 300mg/m(2); 24 received >550mg/m(2). The median follow-up time was 20months. Of the 184 patients, the majority (n=157, 85.3%) did not undergo either an initial cardiac evaluation or surveillance during or post-PLD treatment. Fifty-three patients considered high risk for anthracycline-induced cardiotoxicity underwent CS. Only three patients (1.6%) in the entire cohort developed CHF that was possibly related to PLD treatment; all had significant pre-existing cardiac risk factors. Selective instead of routine use of CS in the study population resulted in a cost savings of $182,552.28. CONCLUSION: Utilizing cardiac surveillance in select women undergoing PLD treatment for gynecologic malignancies resulted in significant health care cost savings without adversely impacting clinical outcomes.

Cost-effectiveness of adjuvant radiotherapy in intermediate risk endometrial cancer.

OBJECTIVES: Endometrial cancer is the most common gynecologic malignancy in the United States. Adjuvant radiotherapy in patients with intermediate risk disease (stage IB, IC, and occult stage II) is controversial. Despite no proven survival advantage, a significant number of women undergo this treatment annually. The purpose of this study was to compare the estimated health and economic outcomes for adjuvant whole pelvic radiotherapy to no treatment with salvage therapy for recurrence. METHODS: A decision analytic model was created to estimate the costs of adjuvant pelvic radiotherapy versus no adjuvant radiotherapy in patients with intermediate risk endometrial cancer. Data used was gathered from published literature and institutional data on costs. The model incorporates complications, recurrence rates, treatment of recurrence, and survival in each group. RESULTS: In the base case analysis, adjuvant pelvic radiation reduced the recurrence rate by 50%. Cost-effectiveness as measured by cost per recurrence prevented was highly sensitive to the probability of recurrence and the efficacy of adjuvant therapy. In our model the mean costs of Strategy 1 with observation and treatment reserved until the time of recurrence would be $5016. In contrast the mean cost of Strategy 2 which incorporated adjuvant radiotherapy would be $21,159. Cost per recurrence prevented based on the incremental cost-effectiveness is thus $225,215. In the highest risk subgroup, using the upper limit of the 90% confidence limit of efficacy seen in GOG Protocol 99, cost/recurrence prevented was approximately $50,000. Results did not differ when using parameters solely from GOG 99 or PORTEC. CONCLUSIONS: Although adjuvant pelvic radiation does not appear to improve survival for intermediate risk endometrial cancer patients, it does prevent recurrences, at a net positive cost compared to no therapy. Data are not currently available to incorporate quality of life information into cost-effectiveness analyses. Obtaining such data would allow cost/quality-adjusted life year gained to be estimated. This information is necessary to determine if the extra costs of adjuvant radiotherapy in patients with intermediate risk endometrial cancer are acceptable by current health care policy standards.

Salvage of isolated vaginal recurrences in women with surgical stage I endometrial cancer: a multiinstitutional experience.

The objective of this study was to evaluate the treatment outcomes and risk factors of women with surgical stage I endometrial adenocarcinoma who were initially treated with surgery alone and subsequently developed isolated vaginal recurrences. Patients with surgical stage I endometrial adenocarcinoma diagnosed from 1975 to 2002 were identified from tumor registry databases at seven institutions. All patients were treated with surgery alone including a total hysterectomy, bilateral salpingo-oophorectomy, pelvic (+/- para-aortic) lymph node dissection, and peritoneal cytology and did not receive postoperative radiation therapy. Vaginal recurrences were documented histologically. Metastatic disease in the chest and abdomen was excluded by radiologic studies. Overall survival was calculated by the Kaplan-Meier method. Sixty-nine women with surgical stage I endometrial cancer with isolated vaginal recurrences were identified. Of the 69 patients, 10 (15%) were diagnosed with stage IA disease, 43 (62%) were diagnosed with stage IB disease, and 16 (23%) were diagnosed with stage IC disease. Patients diagnosed with grade 1 disease were 22 (32%), grade 2 disease were 26 (38%), and grade 3 disease were 21 (30%). Among women, 81% with isolated vaginal recurrences were salvaged with radiation therapy. The mean time to recurrence was 24 months, and the mean follow-up was 63 months. Among women, 18% died from subsequent recurrent disease. The 5-year overall survival was 75%. The majority of isolated vaginal recurrences in women with surgical stage I endometrial cancer can be successfully salvaged with radiation therapy, further questioning the role of adjuvant therapy for patients with uterine-confined endometrial cancer at the time of initial diagnosis.

Comparison of gracilis and rectus abdominis myocutaneous flap neovaginal reconstruction performed during radical pelvic surgery: flap-specific morbidity.

To compare flap-specific complications of gracilis myocutaneous (GM) and rectus abdominis myocutaneous (RAM) flap neovaginal reconstructions after radical pelvic surgery. The study was a single-institution retrospective review of patients undergoing concurrent radical pelvic surgery with GM or RAM neovaginal reconstructions performed on a gynecological oncology service, 1978-2003. Flap-specific complications were compared between the techniques. Forty-four GM and 32 RAM neovaginal reconstructions were analyzed: plastic surgeons developed 12 (27%) GM and 4 (13%) RAM flaps, with all other flaps performed by gynecological oncologists. Primary procedures included 54 (71%) total pelvic exenterations, with partial exenterations or radical vulvovaginectomies in 16 (21%) and 6 (8%) patients, respectively. Forty (53%) patients had received radiation and 28 (36%) received chemoradiation before radical surgery. There were no significant differences in patient characteristics, other than more frequent use of continent urinary conduits (P < 0.001) and a trend for more frequent sidewall radiation (P < 0.1) in the RAM group, reflecting use in more recent patients (P < 0.001). Median follow-up is 28 months (range: 2 weeks to 216 months), with 5% acute operative mortality. Flap-specific complications were significantly increased in GM patients (P < 0.03). Overall flap loss was significantly increased in GM patients (P < 0.02). Thirty (59%) of 51 patients surviving for more than 12 months reported coitus, with no significant difference between the groups. Because of lower overall incidence of flap-specific complications and significantly lower incidence of flap loss compared with GM flap, RAM flap has become our technique of choice for neovaginal reconstruction concurrent with radical pelvic surgery.

Phase I/II trial of intravenous Doxil and whole abdomen hyperthermia in patients with refractory ovarian cancer.

OBJECTIVE: A phase I/II study of Doxil combined with whole abdomen hyperthermia was conducted in patients with refractory ovarian cancer. Liposomal doxorubicin combined with hyperthermia has been shown to increase both liposomal delivery and drug extravasation into tumour xenografts resulting in enhanced cytotoxic effects. PATIENTS AND METHODS: Thirty patients with either recurrent or persistent epithelial ovarian cancer were enrolled. All patients had either measurable or assessable disease. Patients received intravenous (IV) Doxil at a dose of 40 mg m-2 as a 1-h infusion followed by whole abdomen hyperthermia. The phase I portion of the study was performed to determine the maximal tolerated dose (MTD) of hyperthermia. Quality of life (QoL) was performed at baseline, prior to each cycle and every 3 months. Plasma pharmacokinetic studies were performed with the first cycle. RESULTS: Ten patients participated in the phase I portion of the study which demonstrated that the MTD of hyperthermia was 60 min after either average vaginal and rectal temperatures of 40 degrees C had been achieved or after 30 min of power application, whichever was shorter. All 30 patients were either paclitaxel and/or platinum resistant initially or developed resistant disease. The median number of prior chemotherapeutic regimens was three (range 2-8) and six patients had been previously treated with Doxil. There were three partial responses for a response rate of 10% (95% CI: [2%, 27%]) and eight patients (27%; 95% CI: [12%, 46%]) had disease stabilization. The median time to progression or death was 3.4 months (95% CI: [2.6, 5.2]) and the median survival was 10.8 months (95% CI: [8.8, 17.4]). Twelve patients (40%) experienced palmar-plantar erythrodysesthesia (PPE), but only four (13%) experienced grade 3-4 PPE toxicity. Doxil systemic exposure was higher in those with grade 3-4 PPE compared to those with no PPE. None of the patients had grade 3-4 thermal toxicity due to hyperthermia. QoL was not decreased in patients responding to therapy. CONCLUSIONS: Therapy with intravenous Doxil and whole abdomen hyperthermia for patients with platinum/paclitaxel resistant ovarian cancer is feasible and does not negatively impact quality of life.

Rectus abdominis myocutaneous flaps for neovaginal reconstruction after radical pelvic surgery.

The objective of this article is to compare the flap-specific complications associated with vertical (VRAM) and transverse (TRAM) rectus abdominis myocutaneous flap vaginal reconstructions performed during radical pelvic procedures. A retrospective chart review was performed to identify all patients who underwent VRAM and TRAM neovaginal reconstructions performed on the Gynecologic Oncology Service at Duke University Medical Center. Flap-specific complications were compared between the two techniques. From 1988 to 2003, 14 VRAM and 18 TRAM flap neovaginal reconstructions were performed on 32 women during the course of 22 (68%) total pelvic exenterations, 8 (25%) partial exenterations, and 2 (6%) radical vulvovaginectomies. Twenty-eight (88%) patients had been previously treated with radiation therapy or concurrent chemoradiation. Associated procedures included continent urinary conduit in 21 (66%), rectosigmoid reanastomosis in 8 (25%), and intraoperative or postoperative sidewall radiation therapy in 7 (22%) of patients. Overall median survival was 14 months (range: 2-week postoperative death to 65 months), with two (6%) acute postoperative mortalities. Fifteen flap-specific complications occurred in 12 (38%) patients, with no significant differences in flap type. Abdominal wound complications included four (12%) superficial wound separations, while one (3%) patient had a fascial dehiscence associated with complex fistulas that contributed to her death, but no patient developed incisional hernia. One patient each developed > 50% flap loss after TRAM and < 50% flap loss after VRAM flap, respectively. Four (12%) patients developed vaginal stricture or stenosis, two (6%) required percutaneous drainage of pelvic abscess or hematoma, and two (6%) developed rectovaginal fistula. Univariate analysis revealed a trend for increasing flap loss with body mass index > 35 (P = 0.056, Fisher exact two-tailed test), but there were no significant associations with other patient characteristics or flap-specific complications. Thirteen (62%) of 21 patients who survived >12 months reported coitus. Both VRAM and TRAM are reliable techniques for neovaginal reconstructions after radical pelvic surgery and have a similar distribution of flap-specific complications involving the donor and recipient sites.

Loss of expression of the p16 tumor suppressor gene is more frequent in advanced ovarian cancers lacking p53 mutations.

OBJECTIVE: The aim of this study was to test the hypothesis that p53 mutations are less frequent in ovarian cancers with alterations in other genes that regulate G1 progression. METHODS: Expression of G1 stimulatory (cyclins D1 and E, cdk4, Ki67) and inhibitory (p16, Rb, p27, p14) genes was analyzed using Western blots in 84 primary ovarian cancers and seven cell lines of known p53 mutation status. Expression of p16 and Rb also was determined using immunohistochemistry and the p16 gene was examined for homozygous deletions and mutations. RESULTS: Loss of p16 protein was more frequent in ovarian cancers with wild-type p53. All four cell lines with wild-type p53 had lost p16 compared to only one of three with mutant p53 genes. p16 expression was absent in 34% (28/82) of primary ovarian cancers, and this was significantly more common in cases with wild-type p53 (14/28, 50%) compared to those with p53 mutations (14/54, 26%, P = 0.03). Homozygous deletion of the p16 gene was found in cell lines lacking p16, but not in any primary cancers. p16 loss was more common in serous (21/52, 40%) than nonserous cancers (4/23, 17%, P = 0.07). Cases that expressed p16 were more likely to express high levels of Rb (47/55, 85%) than p16-negative cases (12/28, 43%, P < 0.001). Loss of Rb occurred in 5/30 (17%) ovarian cancers lacking p53 mutations compared to 5/54 (9%) cases with p53 mutations (P = 0.48). Expression of G1 stimulatory proteins (cyclins D1 and E, cdk4, Ki67) did not correlate with p53 mutation status. CONCLUSIONS: Loss of expression of the p16 tumor suppressor occurs more often in ovarian cancers lacking p53 mutations. These data are consistent with the paradigm that inactivation of p53 is less of a requisite event in ovarian carcinogenesis when another G1 regulatory gene such as p16 already has been inactivated.

OVX1, macrophage-colony stimulating factor, and CA-125-II as tumor markers for epithelial ovarian carcinoma: a critical appraisal.

BACKGROUND: Ovarian carcinoma remains the leading cause of death from gynecologic malignancy in Australia, the Netherlands, and the United States. CA-125-II, the most widely used serum marker, has limited sensitivity and specificity for detecting small-volume, early-stage disease. Therefore, a panel of three serum tumor markers-OVX1, CA-125-II, and macrophage-colony stimulating factor (M-CSF)-has been used to evaluate the sensitivity and specificity of multiple markers for the detection of early-stage ovarian carcinoma. METHODS: Preoperative serum levels of OVX1, CA-125-II, and M-CSF were measured in 281 patients with primary ovarian epithelial tumors of different histotypes. Among these tumors, 175 were malignant, 29 were of borderline malignancy, and 77 were benign. The three markers also were measured in sera from 117 apparently healthy women. Marker levels were considered abnormal at CA-125-II > 35 U/mL, OVX1 > 7.2 U/mL, and M-CSF > 3.5 ng/mL. RESULTS: Among 175 women with malignant ovarian tumors, at least one of the three serum markers was elevated in 85%, whereas CA-125-II was elevated in 80% (P = 0.008). In 58 patients with Stage I ovarian carcinoma, at least one of the three serum markers was elevated in 76%, whereas CA-125 levels were elevated in 66% (P = 0.04). For patients with borderline and benign tumors, a combination of the three antigens had slightly higher sensitivity compared with CA-125-II, but the differences were not statistically significant. Among 117 apparently healthy women, CA-125-II was elevated in 4%, and one of the three markers was positive in 17%. CONCLUSIONS: The sensitivity of a combination of three serum markers was significantly greater than the sensitivity of the CA-125-II assay alone in patients with primary ovarian epithelial tumors of different histotypes. This was true for all stages, including early-stage, potentially curable disease. When used as single markers, however, only the CA-125-II assay could distinguish invasive Stage I tumors from apparently healthy women.

Relationship between expression of coactivators and corepressors of hormone receptors and resistance of ovarian cancers to growth regulation by steroid hormones.

OBJECTIVE: To determine whether aberrant expression of hormone receptor corepressors or coactivators or defects in estrogen receptor-mediated transcription might underlie resistance of ovarian cancers to hormonal therapy. METHODS: Northern analysis, Western analysis, and polymerase chain reaction were used to examine expression of estrogen receptor (ER), progesterone receptor (PR), the nuclear receptor corepressors N-CoR and SMRT, and the steroid receptor coactivator BRG-1 in ovarian cancer cell lines and primary cancers. The effect of BRG-1 transfection on ER-mediated transcription was examined. We also determined the effect of estrogen and the pure estrogen antagonist ICI 182,780 on cell cycle profile and expression of ER. Finally, we examined the ability of estrogen to upregulate expression of known estrogen-responsive genes. RESULTS: Among primary ovarian cancers, 18 of 52 (35%) expressed N-CoR, and 37 of 52 (71%) expressed SMRT, but there was no correlation between expression of corepressors and hormone receptor status. All of the primary ovarian cancers and cell lines expressed BRG-1. Estrogen stimulation of two cell lines expressing ER (SKOV3, OVCA 432) elicited low levels of ER-mediated transcription that was not enhanced by BRG-1 transfection. ICI 182,780 did not induce cell cycle arrest in these cell lines, but there was evidence of downregulation of ER, indicating a ligand-receptor interaction. However, estrogen did not elicit increased transcription of estrogen-responsive genes (PR, myc, fos, pS2). CONCLUSION: Inappropriate expression of the nuclear corepressors N-CoR and SMRT or the coactivator BRG-1 does not underlie the resistance of ovarian cancers to hormonal therapy. Further studies are needed to elucidate the mechanisms underlying the inability of ovarian cancers to undergo ER-mediated transcription if we hope to understand their resistance to hormonal therapy.

Chemotherapy-induced apoptosis in epithelial ovarian cancers.

OBJECTIVE: To determine whether chemotherapy drugs elicit programmed cell death (apoptosis) in ovarian cancer cells. METHODS: Monolayers of immortalized ovarian cancer cell lines and primary ovarian cancer cells obtained from ascites were grown in the presence of cisplatin, 4-hydroxyperoxy-cyclophosphamide (the active metabolite of cyclophosphamide) or paclitaxel. Next, DNA was extracted from the cells and subjected to electrophoresis to determine if DNA laddering characteristic of apoptosis was present. RESULTS: In three of six immortalized cell lines (OVCA 420, 429, and 433), apoptosis was not seen in response to any of the three drugs. In contrast, in OVCAR-3 and OVCA 432, DNA laddering consistent with apoptosis was observed in response to all three drugs. In the DOV 13 cell line, apoptosis was seen only with 4-hydroxyperoxycyclophosphamide. Among three primary ovarian cancers, cisplatin elicited apoptosis in one case. Both cell lines with mutant p53 genes (OVCAR-3 and OVCA 432) underwent apoptosis in response to all three drugs, whereas among three cell lines known to have normal p53 genes, one underwent apoptosis in response to 4-hydroxyperoxycyclophosphamide and two were unaffected. CONCLUSION: Ovarian cancer cell death in response to commonly used chemotherapeutic drugs involves the induction of a genetically programmed sequence of events (apoptosis) rather than simply necrosis.

Regulation of apoptosis in normal and malignant ovarian epithelial cells by transforming growth factor beta.

Previously, we found that transforming growth factor beta (TGF-beta) inhibits proliferation of normal human ovarian epithelial cells. In addition, although only 1 of 5 immortalized ovarian cancer cell lines was inhibited, TGF-beta inhibited proliferation of 19 of 20 primary epithelial ovarian cancers. In this study, we examined whether TGF-beta induces apoptosis in normal and malignant ovarian epithelial cells. Among 5 immortalized cell lines, only OVCA 420 is markedly growth inhibited by TGF-beta, and this was the only cell line in which TGF-beta elicited DNA fragmentation characteristic of apoptosis. Induction of apoptosis in OVCA 420 was time and concentration dependent and could be partially inhibited by concurrent treatment with an anti-TGF-beta mAb. Although apoptosis was not seen in normal ovarian epithelial cells (n = 7), [3H]thymidine incorporation was inhibited in all cases [mean = 61.2 +/- 7.2% (SD) of untreated control; P < 0.01]. Similarly, TGF-beta inhibited [3H]thymidine incorporation in all 10 primary ovarian cancers (mean = 40.4 +/- 7.1% of control; P < 0.01), but only 3 of 10 (30%) were found to undergo apoptosis when treated with TGF-beta. There was no relationship between p53 status of the ovarian cancers and the ability of TGF-beta to elicit apoptosis. In conclusion, TGF-beta inhibits proliferation but does not induce apoptosis in normal human ovarian epithelial cells. In contrast, some ovarian cancers that are growth inhibited by TGF-beta also undergo apoptosis. These data are consistent with the hypothesis that malignant cells are more susceptible to apoptosis than their normal nontransformed counterparts.