Long-term Cause-specific Mortality Of Adolescent And Young Adult (AYA) Survivors Of Bone And Soft Tissue Sarcoma (STS): A Population-based Study

Reviewer: Abigail Berman Milby, MD
The Abramson Cancer Center of the University of Pennsylvania
Last Modified: October 4, 2011

Presenting Author: P. Youn
Presenter's Affiliation: Department of Radiation Oncology, University of Rochester School of Medicine and Dentistry, Rochester, NY


  • Combined modality therapy has improved outcomes in the treatment of bone and soft tissue sarcomas (STS).
    • The NCI study randomized patients to surgery alone versus surgery plus postoperative radiation with concurrent chemo versus chemotherapy alone. The addition of radiotherapy decreased local recurrence (Yang et al., JCO, 1998).
    • However, the addition of chemotherapy and/or radiation in these patients can affect long-term survival by inducing second malignancies or non-cancer morbidity.
  • Studies in pediatric patients have shown an increase in hematologic malignancies with a short latency (3.1 y) and in solid second tumors with a long latency (11.6 y) (Vasudevan et al., J Surg Res, 2010).
  • Despite overall improvements in cure rates of bone and STS, no large population-based study has evaluated long-term, cause-specific mortality in the adolescent and young adult (AYA, age 15-39) population.

Materials and Methods

  • From the SEER registry, the authors identified 28,844 bone and soft tissue AYA survivors.
    • Patients were classified as having chemosensitive tumors (CTS) such as osteosarcoma, Ewing sarcoma, or rhabdomyosarcoma (RMS) versus non-chemosensitive tumors (NSCT).
    • Patients were then classified by treatment modality: surgery only, chemotherapy +/- surgery (CT), radiotherapy+/- surgery (RT), or CT+RT +/- surgery.
  • Observed numbers of death, standardized mortality ratios (SMRs), and absolute excess risk (AER) for death from second cancer (SC) or non-cancer causes (NC) were calculated to evaluate associations between cause-specific mortality and cancer histology, age at cancer diagnosis, and initial treatment.


  • Patients had the following treatment modalities (percent of CTS and NSCT, respectively): surgery only (15 and 67), chemotherapy +/- surgery (CT) (60 and 6), radiotherapy+/- surgery (RT) (3 and 16), CT+RT +/- surgery (27 and 7), or other (4 and 5).
  • The following percentages of each group died: CTS=24% including osteosarcoma=18%, Ewing sarcoma=30%, and RMS=29%, 10% of NCST died.
  • For CTS, there were 84 non-relapse deaths, with a SMR of 2.8 and AER of 32. There were 28 2nd cancer deaths with SMR of 5.2 and AER of 13 and 56 non-cancer deaths with SMR of 2.2 and AER of 18.
    • By treatment modality:
      • CT +/- surg SMR was 3.2 for non-relapse death, 5.2 for 2nd cancer death, 2.8 for non-cancer death.
      • RT +/- surg SMR was 4.5 for non-relapse death, 9.5 for 2nd cancer death.
      • CT + RT +/- surg SMR was 3.0 for non-relapse death, 11 for 2nd cancer death.
  • For NCTS, there were 379 non-relapse deaths, with a SMR of 1.6 and AER of 17. There were 104 2nd cancer deaths with SMR of 1.8 and AER of 5 and 275 non-cancer deaths with SMR of 1.6 and AER of 12.
    • By treatment modality:
      • Surgery only SMR was 1.28 for non-relapse death, 1.2 for 2nd cancer death, 1.31 for non-cancer death.
      • CT +/- surg SMR was 3.33 for non-relapse death, 6.45 for 2nd cancer death, 2.31 for non-cancer death.
      • RT +/- surg SMR was 2.05 for non-relapse death, 2.52 for 2nd cancer death, 1.9 for non-cancer death.
      • CT + RT +/- surg SMR was 5.05 for non-relapsedeath, 8.69 for 2nd cancer death, 4.1 for non-cancer death.

Author's Conclusions

  • AYA bone and soft tissue sarcoma survivors have an increased risk of death from SC or NC compared to the general AYA population
  • For ES, OS, RMS, non-relapse mortality is more elevated if the patient received chemotherapy or radiotherapy.
  • For other soft tissue sarcomas, non-relapse mortality is elevated in all treatment groups, but only moderately if the patient received surgery only.
  • Further research is needed to develop preventative and surveillance guidelines for both SC and NC diseases to reduce long-term excess mortality among sarcoma survivors.

Clinical Implications

  • The investigators have identified that bone and STS in the adolescent and young adult population have an increased risk of death from secondary cancers and non-cancer causes compared to the general population.
  • This risk was increased even in the surgery only population, although the risk was greater if the patient also received chemotherapy and radiotherapy.
  • It would be interesting for the authors to report what the non-cancer causes of death were, and if these can be easily explained by site of radiotherapy or type of chemotherapy. Likewise, it would be interesting to see the percentage of second malignancies that were hematologic versus solid.
  • It has been proposed that type of radiotherapy could impact the risk of second malignancies; for example, IMRT may increase the risk because of the low-dose bath of radiotherapy and proton therapy may decrease the risk.
    • Yock et al. (ASTRO Abstract, 2008) showed a decrease in the risk of secondary malignancies with proton radiotherapy.
    • Subset analyses examining the 2nd cancer risk in this population based on type of radiotherapy would be interesting, but this is not possible to perform with SEER data.
  • This work emphasizes the need for survivorship care plans that provide close monitoring after treatment for bone and STS in young adults.