Fresh Focus on Second Primary Cancers

OncologyLive, Vol. 19/No. 7, Volume 19, Issue 7

When investigators looked at roughly 150,000 cases of well-differentiated thyroid cancer using the SEER database of the National Cancer Institute, they knew that the use of radioactive iodine in conjunction with surgery and other treatments for thyroid cancer had increased dramatically over the years.

Sudipto Mukherjee, MD, MPH

When investigators looked at roughly 150,000 cases of well-differentiated thyroid cancer (WDTC) using the SEER database of the National Cancer Institute, they knew that the use of radioactive iodine (RAI) in conjunction with surgery and other treatments for thyroid cancer had increased dramatically over the years. This was concerning because RAI is a risk factor in second hematologic malignancies (SHMs), and its use is largely unnecessary in many cases of low-risk WDTC, according to Sudipto Mukherjee, MD, MPH, an associate staff member of the Department of Hematology and Oncology at Cleveland Clinic in Ohio.1 Mukherjee and colleagues sought to determine whether oncologists and hematologists were overtreating with RAI in cases of WDTC, and thus putting their patients at risk for SHMs. The answer was a disheartening “yes.“

Of the 148,215 patients with WDTC, 53% received surgery and 47% received RAI. The RAI treatment was associated with an increased early risk of developing acute myeloid leukemia (AML; HR, 1.79; 95% CI, 1.13-2.82; P = .01) and chronic myeloid leukemia (CML; HR, 3.44; 95% CI, 1.87- 6.36; P <.001). The increased risk of AML and CML after RAI was seen even in the aftermath of low- and intermediate-risk WDTC tumors, which Mukherjee said would not have been the case had RAI been avoided as a treatment option.

“What we have noticed is that there has been a signal of increased second cancers in these patients. Most of them have involved solid tumors, a few have been hematologic malignancies, and in the recognition of these second cancers occurring as a late complication, the American Thyroid Association guidelines have started to emphasize a clinical and judicious application of RAI on a case-by-case basis,” said Mukherjee in an interview with OncologyLive®. The numbers already give hope that clinical practices are changing toward conservative use of RAI, he explained, noting that although use of RAI rose from 6% of cases in 1973 to 50% in 2006, it had declined to 41% by 2014.

For Mukherjee and his fellow investigators, it was important to gain greater insight into the use of RAI and its implications for a broad spectrum of SHMs. Second cancers that result from initial cancer therapy or are caused by other factors are a growing concern, given that cancer survivors are living longer, which makes it important to think about the longer-term consequences of anticancer therapy options. In addition, having individualized survivorship care plans and communicating adequately with patients, primary care providers, and primary care nurse practitioners are vital to ensuring appropriate follow-up care years after an experience of treatment for cancer, said providers interviewed for this article.

“It is projected that by 2020, there will be almost 18 million long-term cancer survivors, and out of that we are expecting probably more than a half a million long-term thyroid cancer survivors,” Mukherjee said. “A substantial majority—probably two-thirds—would have been exposed to RAI, and thus are at risk of AML and CML. That’s a large population who are at risk.”

Concerns about second primary malignancies (SPMs) in patients who have been treated successfully for thyroid cancer are reflected in similar trends in other tumor types. In an analysis of the cancer profiles of patients in SEER registries published in JAMA Oncology in November, investigators found that 18.4% of the 765,843 cancers diagnosed in the study population—or 141,021 malignancies—represented a second order or higher primary cancer.2

Thyroid Cancer Example

Broadly speaking, the National Cancer Institute defines an SPM as a new cancer occurring in a patient with a previous diagnosis of cancer (as opposed to a metastasis). Although the precise reasons for the development of these cancers are not fully understand, investigators have identified 3 major areas of risk: treatment-related causes; genetic susceptibility; and lifestyle, environmental, or medical history factors.3Although thyroid cancers are a fraction of the new malignancies diagnosed annually in the United States, the link between second cancers and prior therapy is a striking example of the implications of prior therapy and the need for early intervention in cases of secondary cancer.

Mukherjee noted that a reduction in the use of RAI is appropriate for patients with newly diagnosed cases of WDTC, but emphasized the value of good follow-up care for early detection and treatment of SHMs. “For people who develop AML following RAI treatment for thyroid cancer, there is an opportunity to save and cure these patients if we are able to detect their cancer early when they are of a relatively younger age, because we can offer them an allogeneic bone marrow transplant as a curative option. In CML, we have excellent therapeutic choices, particularly tyrosine kinase inhibitors [TKIs]. This has now allowed these patients, even after thyroid cancer treated with RAI, to have a normal life span if they respond to TKIs.”

Even so, the use of RAI in low-risk tumors is questionable, and it is better to avoid using this therapy whenever possible, Mukherjee said. Patients with WDTC are, on average, in their mid-40s when they undergo initial treatment, which is usually surgery or surgery followed by RAI, and the vast majority of these patients are still alive 10 years after treatment.

“It would be extremely unfortunate if these patients end up having second, potentially fatal [hematologic malignancies] that cut short their lives because of a treatment that was given in the first place where it was not even necessary or indicated. The purpose of our study was to bring that concern and that risk to the forefront so that people realize that indiscriminate use of RAI, particularly in low-risk tumors, may induce a late complication that is not necessary, because these patients will survive that thyroid cancer and go on to live very normal lives—probably normal life expectancy,” Mukherjee said.

The risk of SPMs in thyroid cancer was partly the focus of a report on SPMs and cancer survivors compiled by Danielle M. Fournier, MSN, AGPCNP-BC, and Angela F. Bazzell, DNP, FNP-BC, from The University of Texas MD Andersen Cancer Center.4 Thyroid cancer survivors have an elevated risk for cancers of the breast, central nervous system, prostate, kidney, stomach, salivary gland, and colon, as well as leukemia, Hodgkin lymphoma (HL), non-Hodgkin lymphoma (NHL), melanoma, and myeloma, the report said, noting that “cumulative incidence of SPM is higher in patients treated with RAI than those who are not.”

The report looked at the incidence of SPMs in cancers with the highest average survival rates, among them breast, melanoma, HL, prostate, and testicular cancers. Over 30 years, the 5-year survival rate of breast cancer in women has increased to 89%, but “hereditary cancer syndromes (BRCA1/2) and exposure to cancer therapies (chemotherapy, radiation therapy, and hormonal therapy) place breast cancer survivors at risk for SPMs,” the report said. “The most common of those cancers include malignancies of the contralateral breast, lung, pleura, esophagus, thyroid, bone, uterus, and ovary, as well as sarcoma and leukemia.”

Second Primary Malignancies Associated With Primary Cancers

Need for Surveillance

In breast cancer, radiation therapy contributes to the development of leukemia as an SPM, and use of the hormone blocker tamoxifen is connected with an elevated risk for the development of endometrial carcinoma and uterine sarcomas. In HL, cancers of the breast and lung are the most common SPMs, but second NHL, melanoma, cervical, thyroid, oropharyngeal, gastrointestinal, and urothelial tract cancers also may result (FIGURE). Again, radiation treatment and the use of alkylating chemotherapy agents are among the suspected causes.Increasingly higher rates of survival for primary cancers make it more important to have guidelines for survivorship surveillance, Fournier and Bazzell wrote, but such guidelines have been established in few specific settings. The American Cancer Society (ACS) and the American Society of Clinical Oncology have developed breast cancer survivorship guidelines that include information on screening for SPMs, and the National Comprehensive Cancer Network has guidelines for HL survivors that provide direction on screening for SPMs. Some guidelines on evaluation for SPMs in prostate cancer survivors have been established by the ACS. Similar guidance is lacking for testicular and thyroid cancer survivors. In melanoma, there are no national screening guidelines for SPMs in survivorship, and neither is there consensus on interventions to prevent and screen for SPMs, Fournier and Bazzell wrote.

Their article, published in the Journal for Nurse Practitioners, was written because they saw a need to educate nurse practitioners (NPs) and other healthcare providers about the risk for SPMs in survivorship care with regard to SPMs. With patients surviving their initial bouts with cancer and moving on to more generalized health maintenance, it follows that non-oncologists should be properly prepared to handle the treatment consequences that may arise. “Even though we addressed NPs’ need for more education for what the risks are for cancer survivors, that can be said for other health professionals, as well,” Bazzell said in an interview. “The survivorship care plan is a good start because it lets the primary care provider—if that information is communicated to them—know that the patient received these drugs or radiation or had this surgery in this time frame, and these are the things they are at risk for.”

One impediment to the success of this strategy is that patients tend to switch providers more often these days, which diminishes the likelihood that their case information will follow them to new care settings and be properly interpreted and acted upon, Fournier and Bazzell said. “There has been more done in the past decade as far as educating survivors about what sort of treatment they received, what are the long-term consequences, and what do they need as far as their follow-up, but the long-term effects of treatment are years if not decades later. And so even if patients are given that knowledge, it’s still something that they forget or aren’t thinking about 10 years later until something comes up unexpectedly,” Bazzell said.

However, malignancies arising from therapy toxicities are just 1 of 3 general contributors to development of second primary cancers. Genetic dispositions and environmental factors, such as smoking, diet, and ultraviolet light exposure, can increase a patient’s risk for a second malignancy. “These are things that the patient can take active responsibility for and make efforts to change, whereas, for instance, patients don’t get to choose their genetic makeup,” Fournier said. Patient education can play an important role in ensuring that cancer survivors are actively involved in this aspect of their health maintenance.

Although Fournier and Bazzell’s report concerned diseases for which treatment success is high, it’s tougher to develop a clear plan of action following cancers for which survival rates are much lower. “As better treatments come out that help extend patient survival for malignancies like lung cancer, how that will impact their SPMs is not something we can fully evaluate if a majority of the patients are deceased at the 5-year point,” Fournier said.

The statistics on lung cancer survival help to illustrate the difficulty of understanding SPMs in hard-to-treat populations. The 5-year survival rates for lung cancer are 15% for men and 21% for women.5 “At advanced stages, the cancer is rarely curable and patients incur significant treatment-related pain and costs with little or no survival benefit. Thus, timely detection in individuals at risk could prevent, interrupt, or delay lung cancer progression,” wrote the authors of a 2015 study on the cost effectiveness of screening for lung cancer.6

The more recent JAMA Oncology study used SEER data to measure the prevalence of prior cancer in patients newly diagnosed with cancer from January 2009 to December 2013.5 Investigators obtained useful data on age groups most likely to have had prior cancer and in what disease types these earlier cancers were most common. They also drew conclusions about the value of adding patients with histories of prior cancer to clinical trial accruals.

In the study, researchers examined the cancer profiles of 740,990 patients from the SEER cancer registries. Incident cancers were divided into 3 groups: first or only primary; second order or higher primary in the same cancer site that occurs after the first cancer; and second order or higher primary in a different cancer site. Of the 765,843 incident cancers diagnosed among that population, 18.4% represented a second order or higher primary cancer. The prevalence of prior cancers differed by age. Among those 20 to 64 years of age, 11% had prior cancer, and in the ≥65 years population, 25.2% did.2

Among those 20 to 64 years of age, prior cancer was most prevalent among incident myeloid and monocytic leukemia; anus, anal canal, and rectum; cervix and other female genital organs; and lung and other respiratory sites (TABLE). Investigators found that among this population, prior cancers generally occurred in different cancer sites, although second order breast, cervical, and other female genital, male genital and testicular cancers were more commonly in the same site.

Table. Prevalence of Prior Cancers Differs by Age Group

In the ≥65 years group, new cancers in patients with prior cancer history were most commonly melanoma; myeloid and monocytic leukemia; bone and joints; and urinary bladder and other urinary organs. Most prior cancers in the older age group were in different sites, with the exception of breast cancer melanoma.

Over 30 years, the cancer survivor population has climbed 4-fold to 15.5 million in 2016 and almost half of all survivors have lived 10 years after their initial diagnosis. They are frequently excluded from cancer clinical trials owing to their prior histories, possibly because it is assumed their disease backgrounds would potentially compromise outcomes, the investigators said. They noted that more than 80% of National Cancer Institute-affiliated lung cancer trials exclude patients with a prior cancer. This is possibly a waste of a valuable resource for generating real-world information.

“One-quarter of older adults (≥65 years) and more than 10% of younger adults newly diagnosed with cancer have a history of prior cancer,” the authors said. They added that the accrual process could be facilitated by including these populations, and the sizeable number of prior cancers among newly diagnosed patients makes it important to study the efficacy of treatments on them, too. Excluding them from trials makes it possible to draw conclusions based only on patients with a first or only primary cancer. “This is partly concerning for older adults with uncommon cancers, where trial accrual is critical, standard therapies may be suboptimal, and prior cancer is prevalent,” the authors wrote.


  1. Molenaar RJ, Sidana S, Radivoyevitch T, et al. Risk of hematologic malignancies after radioiodine treatment of well-differentiated thyroid cancer [published online December 18, 2017]. J Clin Oncol. 2017. Accessed January 14, 2018.
  2. Murphy CC, Gerber DE, Pruitt SL. Prevalence of prior cancer among persons newly diagnosed with cancer: an initial report from the Surveillance, Epidemiology, and End Results Program [published online November 22, 2017]. JAMA Oncol. 2017. doi: 10.1001/jamaoncol.2017.3605.
  3. Second primary cancers. National Cancer Institute Division of Cancer Epidemiology & Genetics. National Cancer Institute website. research/what-we-study/second-cancers. Accessed March 25, 2018.
  4. Fournier DM, Bazzell AF. Second primary malignancies in cancer survivors [published online December 1, 2017]. J Nurse Pract. doi: http://dx. 10.1016/j.nurpra.2017.09.026.
  5. American Cancer Society. Cancer facts & figures 2017. Accessed January 14, 2018.
  6. Puggina A, Broumas A, Ricciardi W, Boccia S. Cost-effectiveness of screening for lung cancer with low-dose computed tomography: a systematic literature review. Eur J Pub Health. 2016;26(1):168-175. doi: 10.1093/eurpub/ckv158.