Secondary cancers after hematopoietic cell transplantation

INTRODUCTION — Hematopoietic cell transplantation (HCT) is an important method for treating certain malignant and nonmalignant disorders. HCT offers potential long-term survival for patients with cancers that have a high risk for relapse, for relapsed or refractory malignancies that cannot be cured using other approaches, and for certain inherited/germline conditions that affect hematopoiesis.

HCT uses chemotherapy and/or radiation therapy as conditioning therapy to reduce the burden of disease and partially or completely ablate hematopoiesis. Hematopoiesis can be restored using autologous cells (ie, the patient's own stored blood or marrow cells) or allogeneic cells (a graft from another individual).

HCT is associated with both short-term and long-term adverse effects (AEs). Short-term AEs include profound cytopenias, infections, toxic effects of the conditioning regimen, and other causes of organ damage. For patients who undergo allogeneic HCT, graft-versus-host disease (GVHD), which is caused by immunologic differences between the transplant recipient (ie, host) and the graft donor, can cause morbidity and transplant-related mortality.

Transplantation is also associated with important late AEs, including secondary cancers (ie, new cancers unrelated to the cause for transplantation), cardiac toxicity, endocrinopathies, and other complications. There is an increase in solid tumors, myeloid malignancies, and lymphoid cancers in the years following HCT. The latency (time until manifestation) of secondary cancers varies with the type of cancer. The increased incidence of secondary cancers is associated with age, aspects of the transplantation technique, GVHD, and other factors.

Transplant recipients should adopt behaviors to lessen cancer risk and have lifelong screening for secondary cancers.

This topic reviews secondary cancers after HCT.

Other late AEs associated with HCT are discussed separately. (See "Long-term care of the adult hematopoietic cell transplantation survivor".)

GENERAL OBSERVATIONS — Patients who undergo HCT are at an increased risk for developing secondary cancers (ie, cancers that are distinct from the condition for which they underwent transplantation).

●Magnitude of risk – Compared with the general population, secondary cancers occur at least twice as frequently in patients who undergo HCT. Secondary cancers can arise after either autologous HCT (in which the patient’s own stem and progenitor cells are used to restore hematopoiesis) or allogeneic HCT (in which the hematopoietic graft is provided by another individual). (See 'Aspects of transplantation' below.)

●Types of secondary cancers – The most common secondary cancers are various solid tumors, myeloid malignancies, and lymphoid proliferations. Secondary solid tumors and myeloid malignancies generally arise years after HCT, while post-transplant proliferative disorder generally occurs within the first year. (See 'Types of second cancers' below.)

●Source of malignant cells in second cancers – Most secondary cancers are derived from the host (ie, transplant recipient) cells. An important exception is post-lymphoproliferative disease, in which the malignancy arises in cells from the transplant donor. (See 'Donor-derived tumors' below.)

Secondary cancers are a significant contributor to nonrelapse mortality (NRM) after HCT, but NRM has decreased significantly over recent decades. Estimated one-year NRM was 30 percent in the 1980s, 24 percent in the 1990s, 15 percent in the 2000s, and 12 percent in the 2010s [1]. Reduced NRM is related to improving supportive care and evolving transplantation technique. However, it is uncertain how newer transplantation techniques, such as nonmyeloablative conditioning or alternative graft donor sources, will affect rates of secondary cancers because there are fewer long-term outcomes data with their use.

HCT survivors are also at risk for other late adverse effects, such as cardiac toxicity, endocrinopathies, and other long-term complications, as described separately. (See "Survival, quality of life, and late complications after hematopoietic cell transplantation in adults".)

RISK FACTORS — Increased risk for secondary cancers is associated with various aspects of transplantation and certain patient characteristics. However, it is difficult to define the risk for a specific individual and to quantify the risk attributable to host factors, the underlying disease, and prior treatments because of the complex clinical history of many transplantation survivors.

Factors that are associated with an increased risk for secondary cancers include [2,3]:

●Aspects of transplantation – Type and intensity of the conditioning regimen, graft source, post-HCT maintenance therapy

●Graft-versus-host disease – Graft-versus-host disease (GVHD) and its management; this is relevant only for patients who undergo allogeneic HCT

●Patient/disease factors – Age at transplantation, current age, underlying disease, lifestyle-related behaviors (eg, smoking, sun exposure)

In addition, some patients have a predisposition toward developing secondary malignancies due to inherited/germline gene variants (mutations) for which they are transplanted (eg, Fanconi anemia [FA]).

Aspects of transplantation — Aspects of the transplantation technique itself contribute to the development of secondary cancers after HCT.

It is uncertain how much the transplant procedure itself contributes relative to previous treatments. Studies that attempted to parse the risk attributable to HCT versus prior therapy reported mixed results and variable effects.

●In one study, there was no difference in the incidence of therapy-related myeloid neoplasms (t-MNs; ie, myelodysplastic syndromes/neoplasms [MDS] and acute myeloid leukemia [AML]) in patients who underwent HCT for Hodgkin lymphoma (HL) compared with matched nontransplanted patients [4].

●A study of aplastic anemia (AA) that included 748 patients who underwent HCT and 860 who were treated with immunosuppressive therapy (IST) reported the 10-year cumulative risk of secondary cancers was 3.1 percent with HCT and 18.8 percent with IST [5]. The distribution of tumors was different in those groups. Although the risk for solid tumors was the same in the two groups, solid tumors accounted for most secondary cancers after HCT, while t-MNs were more common after IST. The major risk factors for malignancy after HCT were increasing age and the use of radiation therapy (RT) in the conditioning regimen.

●One study reported that the risk factors for secondary t-MNs were similar to those associated with difficulty obtaining an adequate stem-cell harvest for autologous HCT; some may interpret this as evidence of an underlying cause for t-MNs that is distinct from the transplant procedure itself. Secondary MDS/AML arose in 6.8 percent of transplanted patients after 10 years, and multivariable analysis reported that prior exposure to RT, ≥4 chemotherapy regimens, and >5 days of apheresis to harvest sufficient hematopoietic stem and progenitor cells were independent risk factors for these secondary cancers [6].

Potential contribution of prior therapies to secondary cancer risk are discussed below. (See 'Underlying disease' below.)

Conditioning regimen — Conditioning therapy uses varying doses of chemotherapy and/or RT to reduce the burden of disease and enable engraftment of transplanted hematopoietic stem/progenitor cells.

Conditioning regimens are described as myeloablative conditioning (MAC), reduced-intensity conditioning (RIC), or nonmyeloablative (NMA) conditioning, depending on the intensity. It is uncertain if either RIC or NMA conditioning is associated with a lower incidence of secondary cancers than MAC in part because there is less long-term follow-up with those techniques. One study attempted to address this and found that after adjustment for other risk factors, there was no difference in cancer risks between RIC/NMA and MAC in leukemia/MDS patients (hazard ratio [HR] 0.98, P = 0.91), but in lymphoma patients, risks were lower in the RIC/NMA group (HR 0.51, P = 0.5) [7]. Further research is needed to more accurately address this issue.

●Radiation therapy – Inclusion of RT in the conditioning regimen is considered an adverse risk factor for secondary cancers after HCT.

In a large registry study, patients who received RT as a component of conditioning therapy had a significantly higher risk of a secondary invasive solid cancer compared with nonirradiated patients (observed/expected [O/E] ratio 2.7 versus 1.3) [8]. Children who received RT at age ≤10 years had a particularly high risk for an invasive solid tumor (O/E ratio 55); the risk remained elevated (O/E ratio 4 to 6) for RT given at age 10 to 29 years, but there was no excess risk evident for those who received RT-based conditioning at ≥30 years [8].

•Total body irradiation – Results have been mixed in studies that assessed risk for secondary cancers when total body irradiation (TBI) is administered. Some studies identified an increased risk from TBI, while others found no such association [9-22]. As above, other factors (age, primary disease, and intensity of prior therapies) likely also factor into the risk for secondary cancers in this setting.

•Age at transplantation – The risk of RT is greater for younger patients than for older patients [8,23].

In one study, among patients irradiated at age <30 years, the relative risk (RR) of nonsquamous cell cancer was nine times that of nonirradiated patients; by contrast, the risk for patients who received RT at age ≥30 years was just 1.1-fold [24].

●Chemotherapy – Incorporation of certain chemotherapy agents into conditioning therapy may affect the risk for secondary cancers.

Among 4318 patients who underwent allogeneic HCT for AML, conditioning with busulfan-cyclophosphamide was associated with a 1.4-fold increased risk compared with the general population [25]. Inclusion of etoposide in the conditioning regimen was not associated with an increased risk for secondary cancers in pediatric patients who underwent autologous HCT [9].

Graft — The graft source, degree of immunologic mismatch, and ex vivo manipulation of the graft may contribute to the risk for secondary cancers, but the extent of these effects is not well defined.

●Graft source – Hematopoietic grafts can be obtained from peripheral blood, bone marrow, or umbilical cord blood (UCB).

•Peripheral blood versus marrow grafts – The incidence of t-MNs is higher in patients who received peripheral blood grafts compared with bone marrow grafts. In a single-institution study, there were 36 cases of MDS or AML among 3372 patients who underwent autologous HCT; compared with bone marrow grafts, the RR for peripheral blood grafts was 3.1 (95% CI 1.3-7.1) [26]. Other studies also reported that the incidence of t-MNs was similar with peripheral blood and bone marrow grafts [10,20,22,27].

•Umbilical cord blood grafts – UCB grafts are associated with an increased risk for certain secondary cancers. However, an accurate assessment of the risk is limited by the small size and relatively short follow-up of most studies. As an example, among 98 adult patients who received UCB grafts, the cumulative incidence of second cancers was 19 percent at two years; post-transplant lymphoproliferative disorders (PTLD) were the most common type of second cancer [19]. (See 'Post-transplant lymphoproliferative disease' below.)

●Human leukocyte antigen match – There is no clear evidence that the degree of immunologic match at human leukocyte antigen (HLA) loci between the graft donor and the transplant recipient affects the risk for secondary cancers after allogeneic HCT.

Grafts can be HLA matched (eg, 8 of 8 or 10 of 10 HLA loci), HLA mismatched (eg, 7 of 8 or 9 of 10 HLA loci), or haploidentical (ie, a graft from a parent or child of the transplant recipient). HLA-matched grafts can come from matched sibling donors (MSD) or matched unrelated donors (MUD); HLA-mismatched grafts can come from unrelated donors, haploidentical relatives, or UCB. (See "Donor selection for hematopoietic cell transplantation".)

•A European Society for Blood and Marrow Transplantation (EBMT) study of 1036 recipients who survived >5 years after transplantation did not find donor-recipient histocompatibility to be a significant risk factor for the development of secondary cancers [28].

•A comparison of 355 MSD and 108 MUD recipients who survived >2 years after transplantation found no effect of donor source on secondary cancer risk [29].

A small percentage of secondary solid tumors and t-MNs are of donor cell origin. An EBMT study of 10,489 allogeneic transplants (1975 to 1998) found 14 cases of donor cell leukemia; 12 were from MSD and 2 were from MUD [30]. However, the small sample size and limited number of unrelated donor grafts limit conclusions regarding the RR for secondary donor-derived cancers. (See 'Donor-derived tumors' below.)

●Graft manipulation – Ex vivo T cell depletion has been associated with an increased risk for PTLD.

The cancer risk associated with T cell depletion appears to be especially elevated when antithymocyte globulin (ATG) or alemtuzumab was used in the conditioning regimen, but there was no difference between MSD versus MUD products [26].

Post-transplant maintenance therapy — The risk of secondary cancers is increased when lenalidomide maintenance is used for multiple myeloma (MM). It is uncertain if other types of maintenance therapy are associated with increased second cancers.

●Lenalidomide – Lenalidomide maintenance therapy after autologous HCT for MM is associated with an increased risk for secondary cancers, but the magnitude of risk varies among studies, and most reports provide limited details of other potential risk factors.

•Meta-analysis – A meta-analysis that included 3254 patients with newly diagnosed MM treated in randomized controlled trials suggested that exposure to lenalidomide plus oral melphalan significantly increased the risk for hematologic malignancies but not for solid tumors [31].

•Randomized trials

-A phase 3 trial reported that, compared with placebo, lenalidomide maintenance therapy after autologous HCT achieved superior progression-free survival (PFS) but was associated with more toxicity and more secondary cancers [32]. Among patients randomly assigned to lenalidomide versus placebo, secondary cancers occurred in 8 percent of patients who received lenalidomide and 3 percent of patients who received placebo.

-In another phase 3 placebo-controlled trial of lenalidomide maintenance therapy after autologous HCT for MM, the incidence of second cancers was 3.1 per 100 patient-years in the lenalidomide group versus 1.2 per 100 patient-years in the placebo group [33].

Further details of these and other studies of maintenance lenalidomide after HCT for MM are discussed separately. (See "Multiple myeloma: Use of hematopoietic cell transplantation", section on 'Maintenance'.)

•Other studies – A Center for International Blood and Marrow Transplant Research registry analysis did not demonstrate an increased risk for second cancers with lenalidomide maintenance therapy [34]. Retrospective single-institution studies reported increased second cancers with lenalidomide maintenance, but the incidence is difficult to interpret because of heterogeneity in patient populations [35,36].

●Other agents – There are no persuasive data that tyrosine kinase inhibitors or other agents used for maintenance therapy after HCT are associated with increased secondary cancers.

Graft-versus-host disease — GVHD is a multisystem condition caused by immunologic differences between the graft donor and the host (transplant recipient) in allogeneic HCT. GVHD does not occur with autologous HCT.

GVHD-related immune dysregulation and treatments to prevent and treat GVHD contribute to immune impairment and an environment that may promote the development of secondary cancers. Allogeneic HCT is followed by a period of lymphopenia and cell-mediated immune deficiency that affects T cells, natural killer (NK) cells, and antibody production. Prophylactic systemic immunosuppression and/or administration of a T cell depleted graft are used to prevent or ameliorate GVHD, and patients who develop GVHD are treated with IST for months or longer.

Patients with chronic GVHD (cGVHD) have at least a two- to threefold increased risk for cancer compared with the general population [3].

●Types of cancer – GVHD is most often associated with secondary squamous cell cancers (SCC), especially those of skin, buccal cavity, and esophagus.

In a study of 17,545 adult recipients of allogeneic HCT, the standardized incidence ratio (SIR) for secondary cancers was 15.7 (95% CI 12.1-20.1); the SIR for esophageal SCC was 8.5 (95% CI 6.1-11.5) [37]. Other studies reported similar findings [24,38,39].

Less is known about risks for other malignancies, although cGVHD has been associated with an increased risk for thyroid cancer [11] and basal cell carcinoma [40].

●Incidence – The risk of SCC was increased threefold in patients with cGVHD, but it was increased nearly 10-fold with severe cGVHD, according to analysis of Center for International Blood and Marrow Transplant Research (CIBMTR) data [24]. Another large study reported the HR for SCC in patients with cGVHD was 3 [40], while a multicenter study described a fivefold increase in risk for SCC in association with cGVHD [8].

●Graft-versus-host disease treatment – Risk for secondary cancers has been associated with the degree and duration of immunosuppression; this is especially notable for SCC of the oral cavity and skin. The impact of specific immunosuppressive agents is less well defined.

•Duration – Longer duration of immunosuppression (ie, the cumulative duration of both prophylaxis and treatment of GVHD) is associated with an increased risk of secondary cancers. Analysis of the CIBMTR database reported that after ≥24 months of immunosuppression, the RR for SCC was 8.4 [24].

•Specific agents – The risk for secondary cancers is increased by certain treatments for cGVHD. Compared with patients who were not treated for cGVHD, the RR for SCC in patients treated with cyclosporine and/or glucocorticoids for cGVHD was not significantly elevated, but when one or both agents were combined with azathioprine, (an agent that is no longer used for immunosuppression after HCT) the RR for SCC was 18.6 [24].

Patient-related factors — Age, sex, the underlying condition (for which HCT is performed), and some behaviors (eg, smoking, sun exposure) can affect the rate of second cancers.

Age — The risk of secondary cancer is associated with the age at transplantation and years since HCT.

●Age at HCT – Children <10 years who received RT as a component of conditioning therapy had a 55-fold increase in the risk of developing a solid cancer, while the risk was four- to sixfold for those irradiated at ages 10 to 29 years; there was no excess risk for those conditioned with RT at ≥30 years [8].

●Time since HCT – Because the risk for solid tumors continues to increase over time, patient age is also associated with an increased risk for secondary cancers.

In a registry study, the O/E ratio for secondary cancers was increased 1.3-fold to 1.6-fold in the first five years after HCT [8]. The O/E ratio increased to 4.6 among survivors of ≥10 years and remained threefold elevated among patients followed for ≥15 years after transplantation.

Sex — There is not a clear difference in secondary cancer rates between males and females.

As an example, a registry study reported a nonsignificant trend toward a higher O/E ratio of secondary cancers in males compared with females (O/E ratio 2.39 versus 1.83, respectively) [8].

Prior treatments — Many patients who undergo HCT have complex medical histories that may have included cytotoxic chemotherapy, RT, or other treatments that can influence the risk for secondary cancers.

●Chemotherapy – Alkylating agents and topoisomerase II inhibitors (eg, anthracyclines, etoposide) are associated with an increased risk for t-MNs, as described separately. (See "Therapy-related myeloid neoplasms: Epidemiology, causes, evaluation, and diagnosis".)

●Radiation – Prior treatment with RT accounts for some of the secondary cancer risk in transplant survivors. Younger children who received RT had a particularly high risk for secondary solid tumors.

•Multivariable analysis of 612 patients who underwent autologous HCT for a lymphoma reported an association of t-MNs with prior RT (ie, RT given before embarking on HCT) but not with prior chemotherapy or with the conditioning regimen; the cumulative incidence of t-MNs six years after HCT was 8.6 percent [12]. A large registry study reported that many patients who developed an invasive solid cancer had previously received RT at or near the site of the subsequent cancer [8].

•A single-institution review of 4905 survivors of allogeneic HCT reported a strong effect of TBI dose on secondary cancer incidence [41]. Compared with chemotherapy-only conditioning, the risk was highest with unfractionated TBI (HR 3.2 [95% CI 1.9-5.3]) or high-dose fractionated TBI (HR 2.1 [95% CI 1.5-3.1]); by contrast, the risk with low-dose TBI (2.0 to 4.5 grey [Gy]) was comparable to that of chemotherapy.

It is unclear if there is a critical exposure threshold for radiation, as there appeared to be a linear relationship between radiation exposure and excess solid cancer risk among survivors of the atomic bombs in Hiroshima and Nagasaki [42]. (See "Approach to the adult survivor of classic Hodgkin lymphoma" and "Acute lymphoblastic leukemia/lymphoblastic lymphoma: Outcomes and late effects of treatment in children and adolescents".)

●Other agents – Pretransplantation treatment with cyclosporine was associated with a fivefold higher rate of secondary cancers among 133 patients with AA who underwent allogeneic HCT [43]. (See "Treatment of acquired aplastic anemia in children and adolescents", section on 'Adverse effects'.)

Lifestyle — Smoking and sun exposure contribute to the risk for secondary cancers, and this may be heightened in recipients of HCT.

Underlying disease — The disease for which HCT was performed may contribute to the risk for secondary cancers. The risk is especially high with transplantation for AA, whether it is acquired or congenital (eg, FA).

Malignancies — It is difficult to quantify the risk attributable to HCT for an underlying cancer because many patients received extensive chemotherapy and/or RT prior to transplantation. Furthermore, many studies are from a time when cancer-associated treatments (eg, nitrogen mustard for initial treatment of HL) were routinely employed.

●Lymphomas – Among patients who underwent autologous HCT for HL and non-Hodgkin lymphoma (NHL), the estimated risk of secondary cancers (primarily MDS and AML) was 8 to 21 percent at 5 to 10 years and as high as 29 percent at 15 years [4,44-48]. Among patients with HL, there was an increased risk for solid tumors (5.2 RR) compared with matched, nontransplanted control patients [4]. The risk was increased sixfold for patients treated with nitrogen mustard, vincristine, procarbazine, and prednisone before transplantation, compared with other regimens. However, some studies reported no association between pre-HCT chemotherapy and the risk of post-transplantation second cancers [10,12]. (See "Second malignancies after treatment of classic Hodgkin lymphoma".)

●Acute myeloid leukemia – Among patients <30 years, a significant excess risk of cancer was noted in patients transplanted for AML (26.6 excess cases per 10,000 patients per year), acute lymphoblastic leukemia (ALL; 17.5 per 10,000 patients per year), and chronic myeloid leukemia (9.7 per 10,000 patients per year) [8]. Among patients ≥30 years, a significant excess risk was noted only in those transplanted for AML (21 per 10,000 patients per year). (See 'Underlying disease' above.)

Marrow Failure Syndromes — Patients with bone marrow failure syndromes are at substantial risk for secondary cancers after HCT. However, both acquired bone marrow failure syndromes (eg, immune-mediated severe aplastic anemia) and congenital/inherited causes (eg, FA) are associated with an increased risk for cancers, even without transplantation.

●A series of 700 patients who were transplanted for marrow failure syndromes reported 14 percent secondary cancers 20 years after allogeneic HCT; the risk was especially high for the 79 patients with FA (42 percent) [49]. The major risk factors were RT as part of the conditioning regimen and use of azathioprine for cGVHD; all but one of the 18 solid tumors were SCC. Secondary solid tumors occurred at a median of 99 months after transplantation; by contrast, the five hematologic malignancies (ALL or PTLD) presented at a median of three months post-transplant.

●Among 157 patients who underwent HCT for FA and survived ≥2 years, the incidence of oral cavity cancers increased over time; the incidence was 8 percent after 10 years and 14 percent after 15 years [50]. SCC accounted for 40 percent of deaths in this series. In a study of 262 patients with FA, the 117 HCT recipients had a fourfold increased risk of developing SCC compared with others who were not transplanted [51].

Risk for the development of secondary malignancies in patients treated with IST (eg, ATG, glucocorticoids, cyclosporine) for acquired AA is discussed separately. (See "Treatment of aplastic anemia in adults".)

Other nonmalignant conditions — It is uncertain if allogeneic HCT for nonmalignant hematologic diseases other than marrow failure syndromes (eg, beta thalassemia, inborn errors of metabolism, immunodeficiency syndromes) are associated with an increased risk for secondary cancers.

In data from the International Bone Marrow Transplant Registry, the incidence of secondary malignancies after HCT for nonmalignant diseases was not increased compared with expected value in the general population; patients with AA and congenital immune deficiencies were excluded from this analysis because they may have an intrinsic increase in the risk for secondary cancers [52].

TYPES OF SECOND CANCERS — There are three categories of secondary cancers after HCT:

●Solid tumors (eg, cancers of skin, oral cavity, breast)

●Myeloid malignancies (eg, myelodysplastic syndromes/neoplasms [MDS] and acute myeloid leukemia [AML])

●Lymphoid tumors (eg, post-transplant lymphoproliferative disorder [PTLD])

The latency period (ie, time for development of a cancer after transplantation) varies among different types of secondary cancers. In general, PTLD develops relatively early, whereas solid tumors generally have a longer latency. Details of the incidence, tumor latency, and causes for each of these categories are discussed in the sections that follow.

Solid tumors — Secondary solid tumors generally begin to appear three to five years after transplantation. The cumulative incidence continues to increase with time since HCT without an apparent plateau. Nearly all types of solid tumors have been described after HCT.

Radiation therapy (RT) and chronic graft-versus-host disease (cGVHD) are major risk factors for secondary solid tumors, but age, underlying disease, prior treatments (eg, chemotherapy and/or RT), smoking, genetic predisposition, and other factors may also contribute.

●Incidence – Most studies report a 1 to 2 percent cumulative incidence of secondary solid tumors after 10 years since HCT; some studies have reported higher rates. The incidence of secondary solid cancers continues to increase over time. Larger studies of secondary solid tumors after allogeneic and/or autologous HCT include:

•In a Center for International Blood and Marrow Transplant Research (CIBMTR) study of 28,874 allogeneic transplant recipients (1964 to 1994), 189 solid cancers were reported [8]. Compared with the general population, the observed/expected (O/E) ratio of secondary cancers was 2.1 (95% CI 1.8-2.5), and the risk increased over time; there was a threefold increased risk among patients followed for ≥10 years after transplantation. The risk of developing a nonsquamous cell cancer (non-SCC) following RT in the conditioning regimen was highly dependent on age at exposure; among those irradiated at <30 years, the relative risk (RR) of non-SCC was nine times that of nonirradiated patients, while the comparable risk for older patients was 1.1. Major determinants of risk for SCC were cGVHD and male sex.

•Among 19,229 patients who underwent allogeneic HCT (1964 to 1992), there was a 2.2 percent cumulative incidence of secondary solid tumors after 10 years; the rate was 6.7 percent after 15 years [52]. The risk was greater for younger patients. The overall RR was 2.7 compared with the general population, and this rose to 8.3 RR for patients who survived >10 years. Total body irradiation (TBI) was associated with higher risk in a multivariate analysis, while SCC of the buccal cavity and skin were strongly linked with cGVHD and male sex.

•A European Cooperative Group for Blood and Marrow Transplantation study of 1036 patients transplanted for leukemia, lymphoma, inborn diseases of hematopoietic and immune systems, or aplastic anemia (AA) reported an incidence of 3.5 percent at 10 years and 12.8 percent at 15 years [28]. Most common were cancers of skin, oral cavity, uterus/cervix, thyroid, breast, and glial tissue.

•In a single-institution study, among 926 consecutive patients who underwent allogeneic HCT (1985 to 2003), the incidence of solid tumors at 10 years was 3.1 percent; these cancers presented after a median of 6.8 years [53]. Compared with the general population, the risk ratio of developing a secondary solid tumor was 1.85; in a multivariate analysis, risk was higher for age >40 years at transplantation and with a female donor.

•A nation-wide study of 17,545 adults in Japan who underwent allogeneic HCT reported that the cumulative incidence of solid tumors was 0.7 percent at 5 years and 1.7 percent at 10 years after transplantation [37]. Extensive cGVHD was a significant risk factor for the development of all solid tumors (RR 1.8), while limited cGVHD was a risk factor for skin cancers.

•Among 2129 patients at a single institution (1976 to 1998) who underwent either autologous or allogeneic HCT for hematologic malignancies, the cumulative incidence of a solid tumor was 6.1 percent at 10 years [10]. The risk was higher for younger patients (ie, age <34 years at the time of HCT).

•The cumulative incidence of solid tumors after autologous HCT (1985 to 2005) in 1347 patients with lymphomas was 2.5 percent at 5 years, 6.8 percent at 10 years, and 9.1 percent at 15 years [54]. The most common solid tumors (among 65 that were detected) were lung, gastrointestinal tract, skin, breast, head and neck, bladder, and thyroid. Patients treated with rituximab had an improved overall survival (OS) but an increased risk for solid tumors (hazard ratio [HR] 0.59 [95% CI 0.47-0.74]); there were nonsignificant trends to increased secondary cancers in association with age >45 and use of RT.

●Risk for specific cancers – The most common secondary solid tumors after HCT include squamous cell cancers (SCCs) of skin, oral cavity, and esophagus; breast cancer; and others. The standardized incidence ratio (SIR; ie, the actual incidence compared with the rate in a control population) is elevated for many solid cancers. In general, TBI is associated with an increased risk for adenocarcinomas, and cGVHD is associated with SCC.

•A CIBMTR analysis of 28,874 children and adults who received allogeneic HCT (1964 to 1994) reported that the greatest increased risk was for cancers of the lip (26.8 SIR), salivary gland (14.2 SIR), tongue (13.3 SIR), bone (8.5 SIR), soft tissue (6.5 SIR), and liver (6.3 SIR) [8]. Most solid tumors increased over time; however, melanomas and thyroid cancers had persistently and stably elevated risk from the first year through ≥10 years after transplantation.

•Among 19,229 patients who underwent allogeneic HCT (1964 to 1992), the risk was most elevated for cancers of bone (13.4 RR), oral cavity (11.1 RR), central nervous system (7.6 RR), hepatocellular (7.5 RR), thyroid (6.6 RR), melanoma (5 RR), and connective tissue (8 RR) [52].

•Among 17,545 Japanese adults who underwent allogeneic HCT, extensive cGVHD was a significant risk factor for the development of all solid tumors (1.8 RR), especially for oral (2.9 RR) and esophageal (5.3 RR) cancers. Limited cGVHD was an independent risk factor for skin cancers (5.8 RR) [37].

•In a single-institution study of 2129 patients who underwent either autologous or allogeneic HCT for hematologic malignancies (1976 to 1998), the SIR was most elevated for cancers of liver (27.7 SIR), oral cavity (17.4 SIR), and cervical (13.3 SIR) [10]. Thyroid, liver, and oral cavity cancers occurred primarily among patients who received TBI.

Other studies have reported similar outcomes [11,15,25,26,53,55,56].

Nonmelanoma skin cancer — HCT survivors are at an increased risk of developing cutaneous SCC and basal cell carcinoma (BCC). RT, GVHD, and age are risk factors for secondary skin cancers.

A study of 4810 patients who underwent allogeneic HCT reported that the cumulative incidence of nonmelanoma skin cancers after 20 years was 6.5 percent for BCC and 3.4 percent for SCC [40]. In a case-control study of 183 patients with post-transplantation secondary solid cancers (58 SCC, 125 non-SCC) and 501 matched control patients (among 24,011 patients who underwent HCT in an international registry), cGVHD and its treatment were strongly related to the risk for SCC; there was no increase in the risk for non-SCC [24].

●Basal cell carcinoma – Compared with nonirradiated patients, the risk of developing a nonsquamous cell carcinoma following RT-based conditioning therapy was increased ninefold for patients transplanted at age <30 years but only 1.1-fold for older patients [8]. In another study, TBI was a significant risk factor for the development of BCC, especially among patients <18 years at HCT; light-skinned patients and those with cGVHD also experienced an increased rate of BCC [40].

Among 3870 patients who received TBI-based conditioning, age-specific rates of BCC were greatest in younger patients; the RR was >20 for those transplanted at age <10 years and decreased with age of exposure until age 40 years, above which there was no excess risk [23].

●Squamous cell cancer – Both acute GVHD (aGVHD) and cGVHD are associated with the development of secondary cutaneous SCC.

Severe cGVHD was associated with a 9.9 RR for SCC; this risk was further increased by the use of azathioprine to treat GVHD [24]. Other studies described an increased risk for SCC in patients with cGVHD, prior chronic lichenoid lesions of the oral mucosa, and/or a history of Fanconi anemia [8,24,25,37]. Post-transplant voriconazole (to prevent fungal infections) has been associated with an increased risk of cutaneous SCC; this has also been seen in solid organ transplant recipients [57].

Oral cavity — SCC of the oral cavity can involve the buccal mucosa, salivary glands, gingiva, lip, or tongue. Oral SCC in this setting is often clinically aggressive. The diagnosis should be suspected in patients with nonhealing oral lesions, leukoplakia, localized oral pain, or changes in the mucosal color or texture.

The source of the malignant cells in four of the eight cases of secondary oral SCC was the donor graft; none of these four transplant recipients had personal risk factors for oral SCC, such as smoking or alcohol overuse, but all had a prior history of extensive cGVHD involving the oral mucosa that required prolonged immunosuppressive therapy [58].

Breast cancer — Major risk factors for secondary breast cancer are younger age at the time of HCT, exposure of breast tissue to RT, and disruption of ovarian function by chemotherapy.

A multicenter study of 3337 female survivors of allogeneic HCT reported 52 cases of breast cancer (cumulative incidence 1.6 percent) at a median of 12.5 years post-transplant; the 25-year cumulative incidence of breast cancer was 11 percent [59]. Breast cancer risk was higher in patients who received TBI (HR 4) and patients <18 years at the time of HCT (HR 9.5).

Myeloid malignancies — HCT survivors are at an increased risk for therapy-related myeloid neoplasms (t-MNs; AML and/or MDS). Risk factors for t-MNs include cytotoxic chemotherapy (eg, alkylating agents) and/or large-field RT that included bone marrow.

Secondary myeloid malignancies should be suspected in patients who develop unexplained cytopenias. Evaluation, diagnosis, treatment, and prognosis of therapy-associated myeloid neoplasms are presented separately. (See "Therapy-related myeloid neoplasms: Epidemiology, causes, evaluation, and diagnosis".)

The risk for t-MNs is elevated after allogeneic HCT, but the rate may be comparable to that of other cancer survivors who were treated with cytotoxic chemotherapeutic agents and RT. The risk for t-MNs may be even higher after autologous HCT because of the substantial cumulative doses received during pretransplantation chemotherapy and radiotherapy, chemotherapy-based stem cell mobilization, and HCT conditioning [12,60,61].

●Retrospective analysis of 1347 patients transplanted for a lymphoma (1985 to 2005) estimated the 5-, 10-, and 15-year cumulative incidences of MDS/acute leukemia were 3.1 percent, 4.5 percent, and 6.8 percent, respectively [54]. Most of the acute leukemias were AML, but there were rare cases of acute lymphoblastic leukemia. The risk was higher among males (HR 3.1 [95% CI 1.52-6.28]) and those who received peripheral blood grafts that were collected during a second harvest (HR 2.2 [95% CI 1.3-4.0]).

●A case-control study of 56 patients with MDS/AML and 168 matched controls (among 2739 patients receiving autologous HCT for lymphomas; 1989 to 1995) reported that the type and intensity of pretransplantation alkylating agent therapy were associated with the greatest risk for t-MNs [60]. In a multivariate analysis, increased risk was associated with the intensity of pretransplantation chemotherapy with chlorambucil or mechlorethamine; there was a trend toward increased risk for leukemia associated with peripheral blood grafts, while TBI ≤12 grey (Gy) was not associated with elevated risk.

●Donor-derived hematologic malignancies have been reported after umbilical cord blood (UCB) transplantation. Despite donor-derived t-MNs in up to 6 percent of transplant recipients of UCB grafts, there have been no reported cases of the same myeloid malignancy in the UCB donor [62]. There are other reports of donor-derived t-MNs after transplantation [63-71]. Similarly, secondary multiple myeloma can occur after allogeneic HCT due to transfer of malignant cells in the graft, but they did not develop in the graft donors [64,69-71].

In addition to t-MNs derived from the transplant recipient’s cells, there are reports of donor-derived t-MNs after HCT. The transplant recipient’s immunologic milieu appears to be critical for the development of t-MNs, since these cancers can arise in transplant recipients through accidental transplantation from an affected donor, despite the absence of the development of the malignancy in the donor.

Post-transplant lymphoproliferative disease — Post-transplant lymphoproliferative disease (PTLD) describes abnormal lymphoid proliferations (usually of B cells) that are derived from lymphoid cells of the transplant donor [72]. Most cases of PTLD arise in the first year after transplantation, and nearly all cases are associated with Epstein-Barr virus. PTLD is almost exclusively seen after allogeneic HCT and is very rare after autologous HCT. The incidence is associated with the degree of immunosuppression, particularly T cell depletion of the graft.

The risk for PTLD is highest within five months of transplantation and declines steeply after one year. The cumulative incidence is generally 1 to 2 percent, but some studies have reported an incidence as high as 8 to 10 percent for patients with multiple risk factors [26,55,73,74].

●Among 26,901 patients who underwent allogeneic HCT (1964 to 1994), PTLD developed in 127 patients; 83 percent arose within one year of transplantation; the O/E ratio for PTLD was 29.7 [73,74]. Factors strongly associated with PTLD were T cell depletion of the donor marrow, use of antithymocyte globulin (ATG), and unrelated or human leukocyte antigen (HLA)-mismatched grafts; risk was also increased in patients with GVHD, age ≥50 years, and second transplantation. The cumulative incidence of PTLD was 0.2 percent among patients with no major risk factors but increased to 1.1, 3.6, and 8.1 percent for patients with 1, 2, and ≥3 major risk factors, respectively [74].

●In a multicenter study of 3182 children (<17 years) who were transplanted for acute leukemia (1964 to 1992), risk factors for PTLD included cGVHD (6.5 RR), unrelated or HLA-disparate related donor (7.5 RR), T cell–depleted graft (4.8 RR), and ATG (3.1 RR) [55].

●In a single-institution study of 3372 transplanted patients, the SIR for non-Hodgkin lymphoma, including PTLD was 54.3 [26]. Factors associated with an increased risk for PTLD included mismatched related donor (9 RR), primary diagnosis of an immune deficiency (2.7 RR), chronic myeloid leukemia (2.5 RR), use of ATG (either in the preparative regimen or for treatment of GVHD; 3.7 RR), T cell depletion (4 RR), and grade ≥3 aGVHD (2.4 RR).

Diagnosis and management of PTLD are discussed in greater detail separately. (See "Epidemiology, clinical manifestations, and diagnosis of post-transplant lymphoproliferative disorders" and "Treatment and prevention of post-transplant lymphoproliferative disorders".)

Donor-derived tumors — The majority of secondary cancers after transplantation are derived from cells of the transplant recipient.

PTLD is an important exception, as this usually arises from donor-derived cells. (See "Epidemiology, clinical manifestations, and diagnosis of post-transplant lymphoproliferative disorders", section on 'Pathogenesis'.)

A subset of t-MNs and solid tumors are of donor origin, as described above. (See 'Myeloid malignancies' above.)

PREVENTION AND SCREENING — All patients who undergo HCT are at an increased risk for secondary cancers and other late effects of transplantation.

Constructing a patient-specific risk profile based on exposures and risk factors is important for developing appropriate screening and prevention strategies for secondary cancers after allogeneic transplantation [75]. In general, health care needs of HCT survivors are best addressed through patient-centered and integrated health care delivery models that coordinate care between providers, institutions, payers, and public health systems.

●Cancer prevention – There is a persistent increase in risk for developing secondary cancers after transplantation. HCT survivors should be advised to adopt behaviors that may reduce the risk for developing a secondary malignancy and encouraged to report any concerning symptoms to their clinicians.

Behaviors that are associated with optimal health for cancer survivors include:

•Smoking cessation – All patients should discontinue smoking to lessen the risk of a secondary cancer. (See "Overview of smoking cessation management in adults".)

•Sun exposure – HCT survivors should avoid unnecessary sun exposure.

•Weight control – A healthy diet and weight control is encouraged.

•Physical activity – Moderate or vigorous activity, as tolerated, is encouraged.

•Alcohol – Limited or no alcohol consumption is encouraged.

Details of these and other prevention strategies are available from the American Cancer Society.

●Cancer screening – Transplanted patients should undergo cancer screening tailored to their individual risk factors. As there is no plateau in incidence of secondary cancers, patients should be counselled regarding the need for life-long surveillance for new malignancies.

Our approach is similar to that proposed in the joint screening recommendations provided by the European Group for Blood and Marrow Transplantation, the Center for International Blood and Marrow Transplant Research, and the American Society for Blood and Marrow Transplantation [76] and by the National Institutes of Health blood and marrow transplant late effects initiative [77].

Routine age-appropriate cancer surveillance is described separately. (See "Overview of preventive care in adults", section on 'Cancer screening'.)

•Clinical assessment – All patients should have a routine physical examination at least annually after autologous or allogeneic HCT.

•Cancer surveillance – Surveillance in transplant survivors should include:

-Skin – Complete skin examination at least annually.

-Mouth – Dental evaluation to monitor for the development of skin and oral cancers at least annually.

More frequent oral examination (eg, every six months) should be considered for patients with chronic graft-versus-host disease, prior chronic lichenoid lesions of the oral mucosa, and/or a history of Fanconi anemia.

-Breast cancer – Female HCT survivors should undergo screening for breast cancer beginning no later than age 40 years. However, there is no consensus method or age of initiation for screening. (See "Approach to the adult survivor of classic Hodgkin lymphoma", section on 'Post-treatment management'.)

An international group recommended that breast cancer screening should begin eight years after radiation or at age 25 years (whichever occurs later) for patients who received total body irradiation or chest irradiation [76].

The American Cancer Society suggested mammography for most patients but proposed that females who received radiation to the chest between ages 10 years and 35 years should undergo both annual breast magnetic resonance imaging (MRI) and mammography [78,79].

-Thyroid – Annual thyroid examination.

-Lung – Annual pulmonary examination with imaging as appropriate.

SUMMARY

●Description – Patients who undergo hematopoietic cell transplantation (HCT) are at an increased risk for secondary cancers (ie, cancers unrelated to the condition for which the patient was transplanted).

HCT uses conditioning therapy (intensive chemotherapy and/or radiation therapy [RT]) to reduce the disease burden, but this partially or completely ablates hematopoiesis unless followed by an infusion of a graft of hematopoietic stem and progenitor cells (HSPCs).

●General observations (see 'General observations' above):

•Types of HCT – Both autologous HCT (the patient’s own stored HSPCs are used to restore hematopoiesis) and allogeneic HCT (a graft from another individual) are associated with secondary cancers.

•Types of secondary cancers – There are increased rates of solid tumors, myeloid malignancies, and lymphoid cancers after HCT.

●Risk factors – Factors associated with an increased risk for secondary cancers include:

•Conditioning therapy – RT and chemotherapy can contribute to the risk. (See 'Conditioning regimen' above.)

•Graft – Use of peripheral blood grafts with autologous HCT. (See 'Graft' above.)

•Post-HCT maintenance therapy – Lenalidomide maintenance therapy. (See 'Post-transplant maintenance therapy' above.)

•Chronic graft-versus-host disease – Chronic graft-versus-host disease (cGVHD) is an important contributor to secondary cancer risk. (See 'Graft-versus-host disease' above.)

•Patient characteristics – Age and smoking affect the incidence of secondary cancers. (See 'Patient-related factors' above.)

•Underlying disease – Risk is highest with HCT for acquired (ie, immune-based) or congenital/inherited aplastic anemia, but it is also elevated in patients transplanted for an underlying malignancy. (See 'Underlying disease' above.)

●Solid tumors – Various types of solid tumors can arise after allogeneic or autologous HCT. Solid tumors are twice as likely in HCT survivors compared with the general population. RT, cGVHD, and other aspects are important risk factors. (See 'Solid tumors' above.)

•Nonmelanoma skin cancers – Cutaneous squamous cell cancers (SCCs) and basal cell cancers are increased. RT and GVHD are major risk factors. (See 'Nonmelanoma skin cancer' above.)

•Oral cavity – There is an increased risk for SCC of the oral cavity. Risk factors include cGVHD, prior chronic lichenoid lesions of oral mucosa, and/or Fanconi anemia. (See 'Oral cavity' above.)

•Breast cancer – Major risk factors for breast cancer include younger age at transplantation, RT that included breast tissue, and disrupted ovarian function (eg, by chemotherapy). (See 'Breast cancer' above.)

●Myeloid malignancies – Acute myeloid leukemia and myelodysplastic syndromes/neoplasms are increased after HCT. Risk factors include cytotoxic chemotherapy (eg, alkylating agents), lenalidomide maintenance, and/or total body irradiation. (See 'Myeloid malignancies' above.)

●Post-transplant lymphoproliferative disease – Post-transplant lymphoproliferative disease occurs within the first year post-HCT and is associated with the presence of Epstein-Barr virus. The risk varies with the degree of immunosuppression, T cell depletion in the graft, and other factors. (See "Epidemiology, clinical manifestations, and diagnosis of post-transplant lymphoproliferative disorders", section on 'Risk factors'.)

●Prevention and screening. (See 'Prevention and screening' above.)

•Increased risk for secondary cancers is lifelong; patients should adopt healthy behaviors.

•Age-appropriate screening should continue throughout life, with adjustment for specific risks.