Changing patterns of practice in the management of nasopharynx carcinoma (NPC): Analysis of the National Cancer Database (NCDB)

Reviewer: Neha Vapiwala, MD
Abramson Cancer Center of the University of Pennsylvania
Last Modified: June 8, 2004

Presenter: Langer, Corey
Presenter's Affiliation: Fox Chase Cancer Center
Type of Session: Scientific

The standard of care in the management of nasopharynx cancer (NPC) has been defined by several randomized phase III trials using concurrent chemoradiation with or without additional adjuvant chemotherapy. Two of these studies (Lin et al from Taiwan and Wee et al from Singapore) included endemic populations, but the US Intergroup report from Al-Sarraf first presented in 1996 is truly the landmark study in this field of research. This trial demonstrated a 40% progression-free and a 30% overall survival improvement with the use of cisplatin-based concurrent chemotherapy and radiation therapy, as compared to radiotherapy alone. The researchers behind the present abstract sought to measure the impact of these data supporting concurrent chemoradiation on actual clinical management of NPC patients in the US. They specifically focused on the effect of Al-Sarraf's data, using the year of its publication in 1998 as an inflection point and comparing practice patterns before and after

Materials and Methods

  • National Cancer Database (NCDB) query was performed for two time periods: 1989-1997 and 1998-2001, roughly representing "before" and "after" Al-Sarraf
  • All patients registered in the database with stage III/IV non-metastatic NPC (and meeting eligibility criteria outlined in the Al-Sarraf study) were selected for the two time periods
  • The use of combined concurrent chemoradiation (CCRT) and patient outcomes were analyzed and compared between the two time periods, particularly looking at the effect of patient age and ethnicity on changing usage patterns, if any
  • A total of 6,327 patients were found for 1989-97:
    • 66% of these were male
    • 63% white, 13% black, 7% Asian
    • 23% stage III, and 77% stage IV
  • A total of 2,315 patients were obtained for 1998-2001:
    • 70% were male
    • 60% white, 16% black, 16.6% Asian
    • 43.4% stage III, and 56.6% stage IV
  • There was no significant difference in age distribution between the two time periods, with a median age of 55-56 years old in both groups.
  • As a general rule, over the years from 1989 to 2001, there has been a progressive increase in the percentage of stage III NPC patients and a progressive decrease in the percentage of stage IV NPC patients. This is partly due to changing staging systems and improving staging techniques over these years.


      • There was a significant increase in the use of CTRT in all NPC patients treated during 1998-2001compared to 1989-97: 65.1% vs. 38.2% overall.
      • There was a near doubling of CTRT use in patients 50-80 years of age between the two time periods.
      • In general, younger patients are more likely to receive CTRT than older patients (~65% in patients less than 30 years old compared to 20% in those over 80 years of age).
      • The percentage of patients aged 16-70 years receiving CTRT during 1998-2001 was 66-74%. 
      • In general, Asian patients are more likely to receive CTRT than white or black patients (81% in Asians vs. 61% in white and black patients, p=0.002).
      • During 1989-98, stage IV patients were more likely to receive CTRT than stage III patients (41.3% vs. 27.6%, p<0.01), but there was no statistically significant difference between the stages during 1998-2001.
      • The 5-year survival data were presented for 1989-92 and 1993-1996, as there is not yet enough follow-up to establish 5-year survival data beyond 1997. The 5-year survival rates are:

      1989-92 1993-96
      Overall (n=5436) 47.8% 50.5%
      Stage III (n=1235) 57.5% 58.5%
      Stage IV (n=4201) 44.7% 48.4%
      White (n=3464) 44.1% 47.1%
      Black (n=689) 52.9% 51.9%
      Asian 59.1% 62.1%
      Age 16-39y 64.9% 68%
      Age 40-59y 49.1% 56.1%
      Age 60+ 36.4% 35.3%
      RT only (n=2157) 45.9% 46.1%
      CTRT (n=1921) 52.6% 56.9%

      Author's Conclusions

      • Clinical management of NPC patients in the US has been postively impacted by the data presented in the Al-Sarraf study on the superior survival gained with CTRT compared to RT alone.
      • Changing practice patterns before and after 1998 reflect this, and there has been an overall increase between the two time periods in the use of CTRT to treat NPC patients as a whole.
      • On subgroup analysis, it appears that younger patients and Asian patients are more likely to receive CTRT than older patients and white/black patients, respectively. 

      Clinical/Scientific Implications
      This patterns of care study is an important follow-up assessment on the impact of clinical data on clinical practice. The growing acceptance of CTRT in the management of NPC patients is apparent in the database query comparing pre- and post-data presentation time periods (namely, 1998). It is interesting to note that younger and Asian patients are more likely to be offered CTRT, as shown in the data here. These patient subgroups also happen to have better corresponding 5-year survival rates as compared to their older and non-Asian counterparts, respectively. While it may not be surprising that younger patients are offered what is typically considered "more aggressive" therapy, it remains to be seen if there is also an effect of younger age on patient outcome, independent of therapy. Unfortunately, this study does not have basic data on the baseline comorbidities of the patients. Similarly, does the patient's ethnicity, specifically Asian heritage, have an impact on long-term outcome that is independent of therapy? It is well recognized that Asian NPC patients have a different histology and a different type of NPC with unique clinical behavior, but this distinction of degree of tumor differentiation is not clearly examined here. Furthermore, this database query has limited utility in that no details are available on the specific treatments (ie: radiation and chemotherpy doses, schedules, etc.) themselves. Nonetheless, several questions for continued research in this area have been raised, and  hopefully these will be answered in the future. In the meantime, this study reassures clinicians and patients that rclinical research data does in fact correlate to clinical reality. 

      Oncolink's ASCO Coverage made possible by an unrestricted Educational Grant from Bristol-Myers Squibb Oncology.