Ipilimumab is the first drug to prolong overall survival in patients with metastatic melanoma and its approval by the US Food and Drug Administration on March 25 made headlines worldwide. Yet, the drug is not straightforward to use. A substantial proportion of patients who eventually respond to the drug initially show tumor growth or new lesions, and side effects must be carefully managed to avoid serious problems. We asked two physicians who participated in ipilimumab clinical trials to share their thoughts on how best to use the drug.

Ipilimumab

Monitoring Response

Unlike conventional cytotoxic chemotherapy, anti-cancer agents that work through the immune system, like ipilimumab, may take time to work. "I think the most important thing for practitioners to understand now, as ipilimumab becomes a commercially available drug that anybody can use, is they shouldn't react to a first disappointing scan the way they would for a patient treated with chemotherapy," said Jedd Wolchok, MD, PhD, Associate Attending Physician at Memorial Sloan-Kettering Cancer Center and a lead investigator on the ipilimumab trials.

He notes that 10 to 25 percent of patients who ultimately benefited from the drug in Phase II trials initially experienced tumor growth or new lesions.

Therefore, patience and clinical judgment are needed. "If a patient is not clinically deteriorating and just the scan is worse, then the suggestion is to repeat the scan in four to six weeks," Dr. Wolchok said. "If on that second scan you see no evidence of things turning around or stabilizing, then you move the patient on to some other therapy."

Dr. Wolchok said he starts preparing his patients emotionally for the possibility of a disappointing first scan right from the start. "Before they even get the IV put it in, I start talking to them about how they will get scans at 12 weeks and that there is a potential that those will look worse but then get better. We just keep reinforcing that."

Immunotherapies do not work fast, agrees Kim Margolin, MD, Professor of Medicine at the University of Washington and Member of the Fred Hutchinson Cancer Research Center. Physicians who haven't used them before will realize pretty quickly, she says, that they have to think of immunotherapies differently than they do cytotoxic agents. Whereas tumor growth over a certain amount during cytotoxic chemotherapy treatment means the agent isn't going to work, that is not always the case with immunotherapies.

So how does a physician know if a particular patient is a delayed responder or a non-responder? "You don't. You absolutely don't," she said. But the period of not knowing still leaves the glimmer of hope, and isn't so bad considering that the alternative is to offer the patient something really toxic, like IL-2, or an unproven experimental agent.

Like Dr. Wolchok, she typically re-images patients after four to six weeks and waits to make a decision until then. If, on the other hand, a patient not only shows signs of tumor progression on scans but is also deteriorating clinically, with a dramatically decreased performance status or labs that are considerably worse than before, then it's probably safe to assume that ipilimumab isn't working and the patient should be treated with something else.

Dr. Margolin, however, notes that because ipilimumab was approved as a short-course therapy, the decision about response will typically be made after the patient has already completed a course of therapy.

"Generally you see imaging results after you have already finished the four treatments at three-week intervals," she said. "That means that most of these decisions are going to come after the patient has completed induction, so you are not talking about continuing the therapy and withholding the next therapy. You are only talking about withholding the next therapy."

Re-induction therapy was not included in the FDA approval and has not been formally tested with the drug, Dr. Margolin noted. During the compassionate use trial, the company, however, provided four additional doses for patients who responded to their initial course but then relapsed. (She likens the re-treatment approach to the intermittent anti-androgen therapy used in prostate cancer patients.) Re-responses were seen with the agent in the compassionate use setting.

As for the question of how or if maintenance therapy should be used with ipilimumab, Dr. Margolin said she expects the question will be addressed in numerous permutations in trial settings since neither re-induction nor maintenance dosing is included in the FDA-approved label. In the clinic, off-protocol, there is no reason to assume patients would do better with maintenance.

"I am a very evidence-based person and until I see the data, I can't say they are missing out on something when we don't give them maintenance," she said.

Managing Toxicity

The document reads in part: "the product can result in severe and fatal immune-mediated adverse reactions due to T-cell activation and proliferation. These immune-mediated reactions may involve any organ system; however, the most common severe immune-mediated adverse reactions are enterocolitis, hepatitis, dermatitis (including toxic epidermal necrolysis), neuropathy, and endocrinopathy. The majority of these immune-mediated reactions initially manifested during treatment; however, a minority occurred weeks to months after discontinuation of Yervoy."

Given these serious side effects, both Drs. Wolchok and Margolin emphasize that regular and early communication between patients and physicians is critical.

"It is all about early communication and not letting things get out of hand," said Dr. Wolchok. "For most of these toxicities, the answer is to give people immunosuppressive medications, such as prednisone. That is not a difficult thing to do."

Dr. Wolchok, however, emphasizes to his patients that if things change, he wants to know about it right away. For example, if a patient starts having three more bowel movements a day than normal, the patient needs to tell him. "I don't care if it is the middle of the night, I want to know about it," he said.

If a patient's bowel movements are just more frequent, then Dr. Wolchok will have his nurse call the patient daily to see how he or she is doing. "If they say it's 15 times a day, watery diarrhea, with crampy pain, then I might do a scan or colonoscopy right away to see how bad the inflammation is, because that is a classic symptom of colitis."

"Usually this doesn't just sneak up on you," Dr. Wolchok said. "The people who present with fulminant symptoms are folks who maybe haven't been telling you about the subtle symptoms for the past week or so."

The other side effects Dr. Wolchok watches out for are liver function abnormalities and pituitary inflammation. He runs tests for liver function before each dose, and may recheck the labs if they are elevated but not high enough to warrant steroid use. For pituitary problems, he wants to know if a patient is more tired than usual, has a drop in appetite, or headaches. In that situation, he is likely to request a brain MRI and thyroid function tests to see how pituitary access is.

Dr. Margolin's strategy is similar. "I give them a very stern sermon about how important it is to let me know what is going on-kind of like after bone marrow transplant," she said. "You need to let me know if you are having diarrhea. You need to let me know if you have a skin rash that is more than a little itching. You need to let me know if you are profoundly fatigued and you can't get out of bed, if you start vomiting, or having other GI symptoms.

"I tell them that," she continued. "I make sure they have all the phone numbers. And I tell them again, because that is the key. Early perception and diagnosis of the complications, and intervention with immunomodulatory agents is how we are going to keep patients out of trouble."

Dr. Wolchok is a paid consultant for Bristol-Myers Squibb, which makes ipilimumab. Dr. Margolin does not have any conflicts to report.

There is little difference in survival or cytogenetic response among younger or older patients with early Philadelphia-positive chronic myeloid leukemia (CML) receiving frontline treatment with imatinib, according to a five-year prospective study of patients enrolled in three clinical trials in Italy.

In an online March 30 advance publication of Blood, the journal of the American Society of Hematology, researchers at several Italian institutes compared data from clinical trials involving 559 patients with early chronic phase CML who were treated for a median of 60 months with imatinib, including 115 who were over age 65 when treatment began. The results showed that while more older patients died during the study period, as would be expected, there was little difference in major molecular or cytogenetic response among patients who began treatment before or after age 60.

The event-free survival rate among older patients was 55%, compared with 67% in their younger counterparts, failure-free survival was 78% vs 92%, progression-free survival was 62% vs 78%, and overall survival was 78% vs 89% in the younger group. But the differences, the researchers found, were all due to higher mortality among the older patients from other causes. No difference was observed when non-CML-related deaths were factored out.

"The major findings of our work were that older patients over the age of 65 had identical response rates-hematologic, cytogenetic, and molecular-as younger patients, and that the mortality related to progression of CML was not different between older and younger patients," said lead researcher Gianantonio Rosti, MD, Professor of Hematology and Oncology at St. Orsola-Malpighi Hospital, at the University of Bologna.

The median age of CML diagnosis is 60 years or older, while most CML patients involved in clinical trials average between 50 and 55, and there has been little data on outcomes in older patients with early-stage CML who start frontline imatinib after age 65. The two largest studies to date have involved fewer patients, with shorter outcomes, or later stage disease, but those studies too found little difference in survival or response between younger and older patients.

"We think, based also on other data in the literature, that a five-year follow-up is sufficient for evaluating response rates and risk of progression. It also represents an important follow-up for evaluation of long-term side effects, both in older and younger patients," Dr. Rosti said. "Of course collection of data regarding the long term safety will go on."

The main limitations of the study, he said, are that it is not possible to estimate the proportion of patients not allocated to imatinib during the enrollment period due to the presence of relevant co-morbidities that generally are more frequent in elderly patients, or due to age. Also, there were no data regarding low-grade toxicities.

"We think that imatinib therapy should be considered for all patients, regardless of age, but given the higher number of comorbidities, more attention in the management of toxicities, dosage, and concomitant drugs is warranted."

'Model' Study

OT Clinical Advisory Editor for Hematology/Oncology Mikkael Sekeres, MD, MS, Director of the Leukemia Program at the Cleveland Clinic Taussig Cancer Institute and Chair of the Hematology/Oncology Pharmacy & Therapeutics Committee, noted that the findings confirm what has been observed in solid tumor patients treated with other therapies.

"Age, in and of itself, does not necessarily equal a poorer response or prognosis to treatment with newer disease modifying drugs. There are differences between patients, but it has become clearer that if you treat older CML patients as aggressively as younger patients, they fare just as well," he said, adding that the new study should serve as a model for other survival studies involving newer molecular-targeted cancer therapies in recognizing that in older patients, there are competing health risks not necessarily related to their cancer.

Nonetheless, getting large study populations for cancers like CML is difficult, as is acquiring accurate cause-of-death data, he explained.

"If a CML patient dies overnight while an internist is on duty and they have to fill out the death certificate, do they write down that the patient died of leukemia or of infection or bleeding? It's a judgment call, and usually leukemia is cited as the cause. Or say a patient is struck by a car and killed while they are crossing the street-Is the cause of death the accident or did they become confused because of an intracerebral hemorrhage due to their disease?"

Usually the cause of death is listed as the accident rather than the disease, he noted.

"Getting accurate cause-of-death data is always tricky, and that makes mortality studies even more challenging.

"What this study does is underscore the importance of our need to revisit and modify existing prognostic systems for evaluating new molecular-based therapeutics to include better mortality data."

MIKKAEL SEKERES, MD, MS, said that the study should serve as a model for other survival studies involving newer molecular-targeted cancer therapies in recognizing that in older patients, there are competing health risks not necessarily related to their cancer.

Earlier Data Confirmed

The findings are similar to those made in 2003 in a smaller study published in Cancer by researchers at the University of Texas MD Anderson Cancer Center (Cortes, Talpaz, O'Brien: Cancer 2003;98: 1105-1113).

In that study, 178 patients with newly diagnosed CML, including 49 patients 60 or older, were treated. The older patients had similar cytogenetic response rates and survival compared with younger patients at a median 16 months of follow-up. Age was not found to be an independent poor prognostic factor for survival or response to treatment.

Lead author Jorge E. Cortes, MD, Professor of Medicine and Chief of the CML Section in the Department of Leukemia, told OT that the new study's longer follow-up period is important, but that the findings come as "no surprise" due to the experience of oncologists treating older CML patients.

"The effectiveness of treatment is already pretty well-known, based a little bit on what is already in the published literature and a lot from perception in the treatment community," he said.

Transplants and interferon used to be the frontline treatments, but not in older patients due to risks of complications and other health issues.

"There has been little published data available about the effectiveness of imatinib as frontline treatment in older patients with early CML, but this new study helps to confirm its effectiveness-it is a straightforward analysis," Dr. Cortes said.

"Data on imatinib in later-stage CML in older patients is much less relevant today. We once used imatinib only after interferon failure, but now we use it at diagnosis. There is no surprise here, but it is always nice to see that confirmation over a longer period has now been published."

Obesity is a well-known risk factor for postmenopausal breast cancer, but the molecular mechanisms behind obesity and an increased risk have been obscure. That situation may now be changing, according to Andrew Dannenberg, MD, speaking during the Joint AACR/ASCO Presidential Forum at the American Association for Cancer Research Annual Meeting.

His team has found, he said, that obesity and excess weight lead to inflammatory lesions in the breast tissue of mice and humans. Macrophages are a major component of these lesions, producing pro-inflammatory cytokines and triggering aromatase activity in mouse models of obesity.

The mechanistic link between excess weight, inflammation, and aromatase reveals potential therapeutic and prevention targets for breast cancer, particularly for hormone-receptor positive tumors. The data may also help explain the link between obesity and triple-negative breast cancer.

"There are many, many examples of chronic inflammatory states predisposing cancer," Dr. Dannenberg, the Henry R. Erle, MD-Roberts Family Professor of Medicine and Director of the Weill Cornell Cancer Center, said in a telephone interview after the meeting. "We have the first evidence that inflammation occurs in the human breast, and it correlates with being overweight or obese, which are known risk factors for the development of postmenopausal breast cancer and also known to be terribly important in terms of negative prognosis for anyone afflicted with breast cancer.

"This discovery is likely to be important for our current understanding and for developing rational interventions to try to attenuate this process," he said. "I think it may have profound public health implications."

Following the Trail

Over the last 10 years, scientists have demonstrated that obesity triggers smoldering inflammation in visceral and subcutaneous fat. (Unlike someone with an arthritic joint, who is likely to report swelling and pain, obese individuals are unlikely to report abdominal pain, despite molecular evidence of inflammation, such as expression of proinflammatory cytokines.)

Based on that observation and the fact that breast tissue is largely composed of fat, Dr. Dannenberg hypothesized that inflammation might be contributing to the increased breast cancer risk in overweight post-menopausal women.

"If you read the tea leaves, as I was reading the tea leaves, if inflammation occurs in visceral and subcutaneous fat, why not the breast?" he said. The difference, of course, is that the breast also has an epithelium, which could respond to the inflammatory signals.

To test the hypothesis, Dr. Dannenberg's team compared the mammary tissue from obese mice that had been fed a high-fat diet with tissue from lean mice fed a low-fat diet. As he predicted, the mammary tissue and visceral fat from obese animals included evidence of inflammation, with the presence of crown-like structures, that were nearly absent in lean animals. The crown-like structures include necrotic adipocytes surrounded by macrophages.

The presence of crown-like structures correlated with increased expression of proinflammatory mediators in the visceral fat and mammary gland, including TNF-[alpha], IL-1[beta], and COX-2. Aromatase expression also increased in the visceral fat and mammary glands from obese animals compared with lean animals, as did aromatase enzymatic activity. Similar results were seen in a genetic mouse model of obesity that does not rely on a high-fat diet.

When the team separated the stromal-vascular fraction of the mammary tissue from the adipose fraction, they found evidence of communication between the two compartments. On the one side, obesity was associated with increased NF-[kappa]B activity in the stromal-vascular fraction, which contains macrophages, leading to increased production of proinflammatory cytokines. Those cytokines then signaled back to the adipose and possibly other cells where they induced aromatase expression and activity. The elevation in aromatase leads to excess estrogen production.

"From our standpoint this is very interesting, because it gives us a number of different molecular hubs that might be targeted through lifestyle, dietary, or pharmacologic strategies to attenuate this phenomenon," Dr. Dannenberg said.

Inflammation in Human Breast Samples

Of course, that strategy will work only if similar inflammatory lesions and molecular signaling occur in women. To find out, Dr. Dannenberg and colleagues examined breast tissue from 30 women who had undergone breast surgery at Memorial Sloan-Kettering Cancer Center. Crown-like structures in breast tissue were detected in seven of 10 overweight women and six of eight obese women, compared with only one of 12 normal weight women. (Dr. Dannenberg said he thinks one reason such structures have not been reported previously is that they are easily missed when tissues are stained with hematoxylin and eosin. By contrast, they are easy to see when the tissue is probed with an antibody against the macrophage-specific marker CD68.)

The percentage of tissue blocks containing crown-like structures also tended to increase with increasing body mass index. However, the correlation is not perfect. Therefore one cannot always predict whether a woman's breast tissue will carry evidence of inflammation based solely on her body mass index.

The team is continuing to work out the molecular aspects of inflammation in human breast tissue, but Dr. Dannenberg said he is pretty convinced that the presence of the crown-like structures indicates that the molecular signaling process will be resemble what was seen in mice.

And given the numerous examples in which chronic inflammation is associated with tumor formation in humans, the chances are good that the breast tissue inflammation is affecting breast cancer risk, he said. "It is my contention that obesity causes an inflammatory process in the breast and will increase the likelihood of human breast cancer and potentially drives progression too."

And there is circumstantial evidence, already, that reducing that inflammation reduces the risk of disease. Dr. Dannenberg and others reported in a 2004 article in the Journal of the American Medical Association that regular aspirin use was associated with a reduction in breast cancer, particularly hormone-receptor positive breast cancer.

At the time, the authors speculated that the effect of aspirin was mediated by inhibiting the production of prostaglandin E₂, a known inducer of aromatase. Consistent with this hypothesis, Dr. Dannenberg found both elevated levels of prostaglandin E₂ and aromatase in the mammary glands of obese mice.

Ray Dubois: Work Very Intriguing

Asked for his opinion for this article, Ray Dubois, MD, PhD, Provost and Executive Vice President at the University of Texas MD Anderson Cancer Center, and an expert on COX-2, inflammation, and colon cancer prevention, said, "This might be an explanation for how obesity contributes to cancer progression. I have always been intrigued by the fact that obesity is a chronic inflammatory disease.

"If you look at inflammatory cytokines and other markers of inflammation, they go up in the blood stream of people with a higher BMI. So obesity does put a stress on the system and sort of revs up the inflammatory response. And we certainly know that obesity increases the risk for some cancers, and breast is certainly one of those.

"Now we need to understand the mechanisms by which obesity does that," Dr. Dubois continued. "If obesity does lead to these local inflammatory lesions, that certainly could set up a situation in the tumor microenvironment that promotes cancer progression. [Dr. Dannenberg's work] is very intriguing, and it could explain how a systemic disease, obesity, contributes to something going on locally in breast tissue."

Dr. Dubois cautions, however, that the number of women studied by Dr. Dannenberg's team, thus far, remains small and so the observations need to be confirmed in a substantially larger population of women.

New Targets & Biomarker Studies

Although Dr. Dannenberg agrees the results need to be confirmed and extended in a larger group of women, he also thinks the observations could be incorporated into ongoing trials as a biomarker for risk. For example, he hypothesizes that studies that examine the impact of adjuvant exercise or weight loss on recurrence risk might find greater benefit in overweight or obese women whose tissue have crown-like structures, relative to women with similar BMI whose tissues don't carry evidence of inflammation.

"Wouldn't you like to know at time zero, if a person has inflammation and how severe it might be, so you can see if they derive bigger benefit?" he said.

"I think the discovery has potentially important implications for future trials and, ultimately, for personalizing therapy, be it behavioral or pharmacologic interventions." Additionally, he notes that the insights might go part way to explaining why aromatase inhibitors appear to be less effective in overweight and obese women, relative to lean women.

Perhaps, most important, though might be the identification of potential targets for intervention, given the growing number of overweight and obese individuals in the population. In addition to the non-steroidal anti-inflammatory drugs that block the COX-2 pathway, there are numerous small molecules and dietary substances that impact the activation of NF-[kappa]B, which resides at the top of a signaling cascade in mammary tissue.

Dr. Dannenberg declined to talk about specifics, because the studies are ongoing, but said that his research group is actively exploring ways to either reduce the number of crown-like structures or render them functionally inert. He predicts either outcome could lead to an improvement for breast cancer prevention or treatment.

"You might say there is cause for optimism," he concluded.

ORLANDO, FL-Women who have among the rarest of cancers-leiomyosarcoma-have far better outcomes when the tumors are excised en bloc than when the cancer undergoes morcellation, usually as part of a laparoscopic procedure. That was the conclusion of a report here at the Meeting on Women's Cancer of the Society of Gyncologic Oncology.

The mortality rate among women whose tumors were removed en bloc during a complete hysterectomy was 19.4% after a mean of 63 months of follow-up while in cases where morcellation-i.e., piecemeal removal of lesions and organs-occurred the mortality rate was 44% after a mean follow-up of 39 months, Jeong-Yeol Park, MD, Clinical Assistant Professor of Medicine at Asan Medical Center in Seoul, said in his plenary talk. "Tumor morcellation and spillage during surgery may adversely affect treatment outcomes in patients with these highly malignant tumors."

In the retrospective study, Dr. Park and colleagues identified 56 patients who underwent surgery that involved removal of leiomyosarcomas: 31 of the women were treated with non-morcellation hysterectomy while morcellation occurred in 25 other women.

The non-morcellation group underwent total abdominal hysterectomy as initial surgery without morcellation; the other women underwent surgery that included abdominal, vaginal, or laparoscopic morcellation.

The researchers narrowed their study population to include just women with early leiomyosarcoma confined to the uterus during surgical management. Also included were patients who were referred to the institution after initial surgery had been performed, and the researchers reviewed the medical records of patients treated between 1989 and 2010.

"We sought to compare treatment outcomes and patterns of recurrence in patients with apparently early uterine leiomyosarcoma who did and did not undergo tumor morcellation during surgery," Dr. Park said.

The surgery often begins as treatment for uterine leiomyoma-fibroids-and advances in minimally invasive surgery may involve morcellation to eliminate the fibroids or perform a hysterectomy with less scarring. There are few symptoms or diagnostic tests that can alert the physician that the "benign lesion" is actually a rare but deadly tumor, Dr. Park explained.

"As a result, many patients with early uterine leiomyosarcoma are diagnosed only after surgical management, which may include tumor morcellation."

When the treating surgeon recognizes that a leiomyosarcoma has been morcellated during surgery, the surgeon often reaches out for help to the gynecologic oncology specialist, but by then fatal damage may have been done. "Once or twice a year we get these phone calls: 'We morcellated a leiomyosarcoma-what should we do next?,'" said Nadeen Abu-Rustum, MD, Director of Minimally Invasive Surgery in the Gynecology Service at Memorial Sloan-Kettering Cancer Center, the Discussant for the study.

"This paper from Korea shows us that morcellating smooth muscle tumors is more problematic for an already very high-risk group of patients. Most of us don't see this tumor at the time of diagnosis. We see it after myomectomy and by that time the patient is referred to us. We know that if there is perforation of the tumor or adjacent soft tissue or there is peritoneal dissemination, these patients do extremely poorly.

"Most of us would agree that if you knew there is a leiomyosarcoma you would not do a morcellation dissection for this tumor. The problem is that with increasing minimally invasive approaches and the benefits of removing big tumors with morcellation this has become very popular as the majority of patients do well and will not have a problem."

Dr. Park, in response to questions from Dr. Abu-Rustum, said that once morcellation of the leiomyosarcoma has been done, no additional workup has been shown to be effective in changing the outcome of the patient.

Spirited Discussion

In a spirited discussion, several doctors expressed concerns over the use of morcellation. "I feel that I am hovering on the brink of safety at all times," said Bobbie Gostout, MD, Chief of Obstetrics and Gynecology at the Mayo Clinic.

She said that gynecologic oncologists have advocated careful removal of suspicious lesions because they understand that missteps can change the women's prognosis. "But we are also the group that seems to have accepted morcellation. As a surgical tool I don't think there is an acceptable, safe morcellator out there.

"That's different than saying "Are we getting away with it? We are getting away with it most of the time. I think it is time to go back to out industry partners and say we need a new alternative. We need a contained system so that we can advance the goal of minimally invasive surgery, which I fully embrace.

"We are exposing our patients to a risk that to me seems out of bounds," Dr. Goustout said.

Session moderator Pedro Ramirez, MD, Director of Minimally Invasive Surgical Research and Education and Associate Professor in the Department of Gynecologic Oncology at the University of Texas MD Anderson Cancer Center, agreed: "That is certainly a very valid comment. At our institution we don't own a morcellator so those cases are not performed."

Dr. Abu-Rustum noted that in attending other societies' meetings that are associated with gynecology, "you can't go to a session without seeing a morcellator for a fibroid or uterus or a new technique. It is so widely accepted, and you get away with it most of the time, but that doesn't make it right."

He suggested that doctors devise an algorithm that can give surgeons an idea of what the risk is, based on various imaging and clinical signs so they can determine if a woman is at high risk for having a malignancy.

"I think the challenge for gynecologic oncologists is that we get referral from gynecologists who say that they have seen only one such case in their lifetime, and 'it doesn't matter anyway because the patient is going to die,'" said Peter Lim, MD, Medical Director for Robotic Surgery and Minimally Invasive Surgery at Center of Hope in Reno, Nevada.

"That is typical of the comments I have gotten. As we do more and more robotic and minimally invasive surgery, I think we are going to see a bigger population.

"We have to identify these patients," Dr. Lim continued. "We have to better refine our tests. We can't just say, 'Well, we don't hava a test. Let's just go ahead and do it.'"

Morcellation is likely to increase as minimally invasive surgery goes toward smaller and smaller incisions-"You have to morcellate, because you can't get a big piece out," he explained.

He noted that when a surgeon morcellates a leiomyosarcoma, "you have Stage 1 disease that is now Stage 4 disease." Morcellating these tumors can create a "tumor seeding" that becomes even more difficult because there are no chemotherapy treatments that are known to work in the disease.

"Fibroids are very common, and the estimated incidence of leiomyosarcoma in fibroids is one in 200,000," said Dr. Gostout. "Minimally invasive surgery has made such a difference for the thousands of women each year who require this surgery. Right now the tools to take out the uterus involve morcellation of some sort, and these tools are just inadequate."

The patients in Dr. Park's study were about the same age-47.9 in the non-morcellation group and 46.4 in the morcellation group. Patients had an average of two children; 33% of the non-morcellation group were menopausal compared with 16% of the morcellation group. One of the non-morcellation patients required reoperation compared with six patients who had morcellation.

More than 90% of the patients were diagnosed with Stage1 disease; the tumors were about 9.8 cm in the non-morcellation patients and 7.3 cm in the morcellated group, and about 905 of the tumors were found to be high grade.

About 60% of the patients in the study received chemotherapy or chemoradiation therapy. The rest of the women did not receive adjuvant therapy.

The five-year overall survival among the non-morcellation patients was 73% compared with 46% for patients in which morcellation occurred.

"Not only complete excision of the tumor but also an en bloc excision without tumor injury is important in the management of patients with apparently early uterine leiomyosarcoma," he said.