Nora Ephron's final act played out in Manhattan on June 26 where the 71-year-old writer and movie director died from pneumonia brought on by acute myeloid leukemia (AML), one of the most common types of leukemia among adults. AML is a cancer caused when abnormal cells grow inside bone marrow and interfere with the production of healthy blood cells. The marrow eventually stops working correctly, leaving a person with an increased risk of bleeding and infections.

Ephron—best known for writing When Harry Met Sally and Sleepless in Seattle—was first diagnosed in 2006 with one of the myelodysplastic syndromes (MDS), a category of blood diseases also referred to as "preleukemia" that can progress into AML if the bone marrow continually fails to produce enough healthy platelets, red blood cells and white blood cells over time. MDS made headlines recently when ABC's Good Morning America anchor Robin Roberts announced she has been diagnosed with the disease.

Some types of leukemia, including AML, develop as a result of exposure to certain chemicals (including herbicides and pesticides), chemotherapy drugs (such as etoposide and a class of drugs known as alkylating agents) and radiation. Typically, however, a doctor is unable to pinpoint the exact cause in individual cases.

Although estimates vary, there are between 10,000 and 12,000 new cases of MDS in the U.S. annually. More than 80 percent of all MDS patients are older than 60. The National Cancer Institute projects that 13,780 men and women—7,350 men and 6,430 women—will be diagnosed with AML and that 10,200 men and women will die of the malady this year.

Scientific American spoke with Bart Scott, a medical oncologist specializing in the treatment of patients with MDS, about syndrome's progression to AML, who is most at risk for this cancer and whether there are any promising treatments on the horizon. Scott is also director of hematology and hematologic malignancies at the Seattle Cancer Care Alliance and an assistant member of the Fred Hutchinson Cancer Research Center's clinical research division.

[An edited transcript of the interview follows.]

What is the typical trajectory for this form of cancer?
The trajectory of myelodysplastic syndrome is highly dependent upon two factors—the bone marrow myeloblast count and the cytogenetics. [A myeloblast is an immature blood cell that will eventually develop into a type of white blood cell.] The higher the bone marrow myeloblast count is at time of diagnosis the more likely a patient is to progress to acute myeloid leukemia. In general, the more abnormal the cytogenetics [a cell's hereditary and functionalm characteristics] are the more likely a patient is to progress to AML, but there are certain abnormalities that have a good prognosis.

Who are the typical victims of this form of leukemia?
The median age of diagnosis of MDS is 72, it is slightly more common in men than women, and there is a correlation between incidence and age. I would say that Ephron's course does seem consistent with MDS progression to AML.

What are the risk factors for this cancer?
They are based on increasing age as well as certain environmental exposures like pesticides and herbicides. This is not casual contact—we are talking about prolonged, persistent exposure. The amount of exposure required to cause the disease is unknown, however.

Given that Ephron lived most of her life in New York City, whose residents are not typically exposed to inordinately large amounts of pesticides or herbicides, what might account for her condition?
Some researchers have tried to understand MDS in terms of clustering, although there is very little clustering data available. The idea is to understand whether MDS tends to cluster in areas of higher exposure to certain toxic agents. In 2007 Xiaomei Ma [an associate professor of epidemiology the Yale School of Public Health] and a team of researchers published a paper in Leukemia Research regarding this phenomenon. They found that cases of MDS in Connecticut were clustered near the western border. In general, clustering supports the idea that environmental exposure is contributory to the development of a disease. This is the only paper that I know of that has demonstrated clustering in MDS.

We are doing a similar study in the state of Washington. Hanford Site is a nuclear production complex near the Tri-Cities area in southeastern Washington [encompassing Kennewick, Pasco and Richland]. We are looking to see if there is clustering in the Tri-Cities area. But these studies are difficult given the mobility of the area's population, unknown exposure time and any delay in the development of MDS after exposure.

What is the usual treatment?
There are three FDA approved treatments. One is azacitidine, approved by the FDA in 2004 and marketed as Vidaza. Decitabine, sold as Dacogen and approved in 2010 to treat MDS, is another option. [The FDA's oncologic drugs advisory panel in February, however, recommended against the approval of Dacogen to treat older patients with AML. The drug's maker wants FDA approval to use Dacogen in AML patients 65 years and older and who are not considered good candidates for high-dose chemotherapy as an initial treatment for AML.] The third is lenalidomide, also known as Revlimid and introduced in 2004.

However, the only curative treatment is stem cell transplantation. Stem cells are infused just like a blood transfusion. Sources of stem cells include cord blood cells, peripheral blood mobilized stem cells or bone marrow. Success depends on the stage of disease. The one-year treatment related mortality is approximately 20 percent. If patients are transplanted in an early stage of the disease, we have a success rate of 80 percent.

Are there any promising treatments in late-stage testing?
Yes, there are several drugs being investigated in phase II and phase III trials. We currently have a phase III trial open with a drug called rigosertib for patients who have failed prior treatment with azacitidine or decitabine, which are known as demethylating agents. The proposed mechanism of action for these drugs is that they alter gene expression profiles in the cancer cells and increase their susceptibility to death. Rigosertib is a cell cycle inhibitor and would prevent the cancer cells from growing and induce direct damage to the cancer cells, causing their death.