Wednesday, February 25, 2009

CT vs CTRT before surgery for Esophageal Cancer

In this week's JCO there is an important article comparing two approaches for esophageal cancer.

Initially presented at ASCO 2007, this is a German trial from Stahl, randomizing uT3-4NXM0 adenocarcinoma of the lower esophagus or gastric cardia to induction chemotherapy (CDDP+5FU+leucovorin x 15 weeks) followed by surgery (arm A) vs chemotherapy (same x 12 weeks) followed by chemoradiotherapy (30Gy @ 2Gy/day with concurrent CDDP+etoposide) followed by surgery. 

Power calculation: Primary outcome was overall survival time, and 354 patients were needed to detect anincrease in 3-year survival from 25% to 35% by addition of radiation therapy.  Unfortunately the trial failed to meet accrual goals and closed.

Results: Median f/u 4years. CTRT associated with increased pCR rates (15.6% v 2.0%) and pN0 disease (64.4% v 37.7%).  3-year survival rate from 27.7% to 47.4% (p =0.07).  Postoperative mortality 10.2% CTRT v 3.8% CT (p=0.26).

This is an important trial as there are now competing philosophies in the treatment of gastric and esophageal adenocarcinoma, perioperative chemotherapy as in the MAGIC trial, and preoperative CTRT as in the CALGB approach.  This is an underpowered trial with a very provocative survival result.  Though stastically insignificant, this difference would be extremely clinically significant.

Monday, February 23, 2009

Treatment Approaches for High Risk Prostate Cancer

Junzo Chino M.D.


  • 55yo AAM presents to Primary Care for initial visit
  • He has mild HTN on exam
  • No GU symptoms
  • Erectile function intact
  • PSA is sent as a screening test and returns at 325 ng/ml
  • DRE performed revealing a moderate sized gland, firm nodule in the R apex.
  • U/S guided Biopsy performed
  • 7/8 cores reveal prostate cancer 4+4 = 8 GS
  • Pt is initially seen in Urology Clinic and offered surgery - placed on OR schedule

Further Staging

  • CT scan reveals moderate sized prostate
  • 3x4cm R pelvic lymph node just below the bifurcation of the common iliac
  • No other LAD
  • No invasion of bladder or rectum
  • Bone scan negative


  • We are called by the urology chief resident, to have the patient discuss alternatives to surgery

Defining High Risk

D'Amico classification looking at T stage, GS and PSA: any of the below

  • T2c or greater
  • GS 8 or greater
  • PSA 20 or greater

Per Nguyen, IJROBP 2009: Increased # of high risk factors is associated with PCSM. Paper includes PSA velocity as an additional risk factor.


Surgery for High Risk Disease

Several Retrospectives series exist comparing results of surgery to radiation (see Zincke Cancer 1981, D'Amico JAMA 1998)

Duke Data (unpublished)

from 1997-2007, 146 patients with clinically high risk disease operated on

  • Median PSA 9
  • 58% had GS 8 or greater
  • 23% had T-stage T2c or greater
  • 91% had only one high risk factor

Therefore this cohort tended to have a high grade and low-middle disease burden

Pathologic Results

  • 76% had at least on indication for adjuvant RT on final path (ECE, SVI, positive margin, or PSA ≥ 0.2ng/ml)
  • 55% with ECE
  • 27% with SVI
  • 54% with positive margins
  • 20% with PSA ≥ 0.2ng/ml
  • 28% received early RT (within first 6 months postoperatively)

Outcomes at 4 years

  • 48% (95% CI 38-57%) had received postoperative radiation therapy
  • 74% (95% CI 64-83%) remained free of biochemical failure (including those salvaged with RT)
  • 98% (95% CI 96-100%) alive
  • 14% (95% CI 8-23%) undergoing indefinite ADT

Postoperative Toxicity

  • ≥ Grade 2 urinary toxicity in 42 (29%)
  • ≥ Grade 2 erectile dysfunction in 93 (64%)
  • ≥ Grade 2 rectal toxicity in 3 (2%).

Conclusions -

  1. Even in Highly selected patients, ½ of patients undergoing initial surgery had postoperative RT by 4 years.
  2. With selected adjuvant RT and salvage RT, 4 year biochemical control was reasonable at 74%
  3. Toxicity rates are not insignificant

New directions for Surgery and Radation Therapy

Neoadjuvant RT for high Risk Prostate cancer

  • Phase I/II trial is proceeding at Duke with reasonable toxicity to date


To Compare Approaches (Apples and Oranges)

Merglen Arch Intern Med 2007;167(18):1944

Compared results with 844 consequtive patients treated for ACP in Geneva Switzerland 1989-1998. This demonstrates high survival rates for those treated with RP vs RT, However, there were clearly significant selection biases for surgery have younger age, and a bias against RT + ADT for higher grade disease. When these factors were controlled for with a Cox Regression, the HR of RT + ADT overlapped the Surgery Cohort.

Moreover, there are multiple other factors that go into Surgical selection that are unaccounted for. No details on RT or ADT - given time frame likely these were suboptimal


SEER data (Lu-Yao Lancet 1997)

queried the SEER databaes 1983-1992

59,876 pts identified

Cohorts determined by reported Grade (1-3) and Therapy Received

No control for PSA, cT stage or comorbity

RT fared worse with G3-4 disease compared to RP, but again there is no systematic control for other variables.


External Beam Radiation Therapy and Hormone Treatment

The majority of randomized data in Prostate cancer treatment lies in this group.


  1. RTOG 8531 (non bulky)
  2. RTOG 8610 (bulky)
  3. RTOG 9202 (4 months vs 2 years ADT)
  4. EORTC 22863 (no ADT vs 3 years ADT)
  5. EORTC 22961 (6mo ADT vs 3years ADT)

RTOG 85-31 "HIgh Risk - NonBulky"

977 patients with cT3 or pT3 ACP, or N+. bulky tumors excluded.

Randomized to

  • RT alone (44-46 WPRT + boost to 65-70Gy)
  • RT + adjuvant goserelin indefinitely

10 year results:

  • OS 49% ADT vs 39% (SS)
  • LF 23% ADT vs 38% (SS)
  • DM 24% vs 39% (SS)

RTOG 86-10 "High Risk Bulky"

471 pts with T2-T4 ACP, bulky randomized to

  • RT alone (45Gy pelvis, boost to 65-70 Gy)
  • goserelin + flutamide x 4months (neoadjuvant and concurrent) + RT

10 year results:

  • OS ADT 43% vs 34% (NS) (primary endpoint)
  • DSM ADT 23% vs 36% (SS)
  • DM ADT 35% vs 47% (SS)
  • BR ADT 65% vs 80% (SS)

RTOG 92-02: Short vs Long ADT

1554 pts with T2c-T4 ACP, psa < 150 Randomized to:

  • Goserelin + Flutamide x 4 months (neoadjuvant and concurrent) + RT WPRT 45Gy boost to 65-70Gy
  • Same + 2 years goserelin thereafter

10 year results

  • OS 52% LTADT vs 54% STADT (NS)
  • GS 8-10: OS 45% LTADT vs 32% STADT (SS)
  • DFS 22% LTADT vs 13% STADT (SS)

EORTC 22863: Bolla

415 pts with t1-2 G3, or T3-4 any grade (majority T3), randomized to

  • RT 50Gy WPRT, boost to 70Gy
  • Same RT + goserelin x 3 years (start with RT)

5 year results

  • OS ADT 78% vs 62% (SS)
  • DSS ADT 94% vs 79% (SS)
  • DFS ADT 74% vs 40% (SS)

D'Amico Study

206 men with Intermediate-High Risk - T1-T2b PSA>= 10, GS >=10, ECE or T3 by MRI

  • 70 GY RT (no pelvis)
  • Same + 6 months leuprolide/goserelin + flutamide (start 2 months prior to RT)

8 year results

  • OS 74% (ADT) vs 61% (no ADT) SS
  • Difference seen primarily in men without significant comorbidities

What about 6 months vs 3 years? ADT: Bolla EORTC 22961

970 pts with pT2c-T4 or N+ prostate cancer PSA <150, randomized to

  • 70Gy + 6 months ADT
  • 70Gy + 3 years ADT

at 5.2 years f/u interim analysis demonstrated futility. 5y OS was 85% LTADT, 81% STADT, failing to prove non-inferiority. 5y bRFS was 78% vs 59%, which was SS for inferiority of STADT


Currently Three Trials exist on +/- WPRT

  1. RTOG 77-06
  2. RTOG 94-13
  3. GETUG-01

None of these trials are perfect

None of these trials demonstrate a benefit to WPRT

However, the EORTC trials and RTOG trials demonstrating benefit of ADT used WPRT


ADT summary table







yrs fu






c or p T3, or N+


∞ A


if N+








bulky T2-4


4mo NA/C

65- 70Gy








EORTC 22863


T1-2 G3 or T3-4


3yr C/A





74% *







4mo NA/C

2yr NA/C/A












6mo NA/C/A








A- Adjuvant, C- Concurrent, NA- Neoadjuvant, * SS, † SS for GS2-6, ‡ SS for GS8-10


At this point I ran out of time, but wanted to cover EBRT with brachy boost as another RT based option which has had good preliminary results - Stay tuned!


Case Resolution

Pt started on ADT

To be rescanned for potential RT + LTADT in 2 months


Friday, February 20, 2009

Adjuvant RT for prostate cancer (SWOG update Journal Club)

AC presents today on the recent SWOG adjuvant RT for prostate cancer update.

Approximately 1/3 of patients undergo prostatectomy for prostate cancer, and of those 1/3 have some form of biochemical recurrence.

There is some debate about the definitions of salvage vs adjuvant RT.  We define adjuvant RT as treatment postoperatively without a detectable PSA, however some institutions will include any treatment within six months of surgery as adjuvant.

30-60% of those undergoing RP have either ECE or Posititive margins.  Additional factors associated with failure include SVI, high GS, and pretreatment PSA.

Randomized Data
EORTC 22911 (Bolla, Lancet 2005) 1005 pts with pT3 or positive margins, randomized to 60Gy vs wait and see.  Results: improved cPFS, LRF, no difference in OS.  See prior post on this trial.  Approximately 2/3 of the wait and see group recieved salvage RT.  Additionally only 10% had detectable PSA.

ARO 96-02 (Wiegel, ASCO 2005) 385 with pT3N0 randomized to 60Gy vs wait and see.  In the randomization, all had an undetectable PSA. 20% had detectable PSA in this trial and were all treated with 66Gy.  in all groups biochemical control improved HR 0.79 p=0.08.  In the adjuvant group only, biochemical control improvement HR 0.49 p=0.001.  See also prior post for link.

SWOG 8794.  pT3 or positive margins, n=425.  Randomized to 60-64Gy vs observation.  Primary endpoint Metastasis Free Survival.  Accrual goal was 558.  Power 0.8, alpha 0.05, HR for MFS would be 0.66.  The accrual goal was revised when a lower event rate was seeing in the control group in 1996.  Longer follow up assumed, and accrual revised to 408.  Also revised to a two-sided test.

10% in intervention arm didn't receive treatment.  ~1/3 of patients in control arm recieved RT.  Also 1/3 had a detectable PSA.

Initial report in JAMA 2006 failed to meet MFS endpoint, though this was borderline p=0.07.  bRFS was improved.

Update with 12.6 years f/u.  DM free survival improved with RT (HR 0.71; 95% CI 0.54, 0.94; p = 0.016), and OS improved with RT (HR 0.72; 95% CI 0.55, 0.96; p = 0.023). NNT was 12 for DM free survival, and for OS NNT was 9.

No subgroup seemed to benefit more from RT.

Moinpour JCO 26:112, 2008: QOL data from this trial.  After year 2 bowel movement QOL was similar between RT vs obs.  In urinary QOL, RT had a consistent reduciton in QOL.  Global QOL seemed to be initially worse in the RT arm, but curves crossed at 2 years, and RT became superior at longer followup.

Thursday, February 19, 2009

Prostascint Scans

MP presents on the utility (or lack thereof) of Prostascint

Prostascint is based off of a PSMA (prostate specific membrane antigen) antibody - capromab.  FDA approved in 1996 for diagnosis and staging in high risk disease and for PSA recurrence after RP.

The scan is performed by In-111 conjugated capromab.  Scan is performed with Tc99m to differentiate pathologic uptake from vasculature.  Recently these have been fused with CT or MRI.  Cost per scan is ~2,000$

Pt presented with intermediate risk disease (GS 3+4 with organ confined disease treated with prostatectomy, with persistent PSA of 0.3ng/ml postoperatively).  A prostascint is obtained and reveals paraaortic uptake without CT correlate, and no activity in the prostate bed.

Test Characteristics.
Sensitivity - true positives/true positives + false negatives
Specificity - true negatives/true negatives + false positives

Elgamal - 100pts with 136 scans.  124 has positives prostascints - all positives were biopsied, and some of the negatives (though this was not well controlled)
Sensitivity - 89%
Specificity - 67%
Bone scan didn't correlate well the Prostascint (sensitivity of prostascint was only 57% for bone scan positive disease)

Multiple studies exist demonstrating a wide range of sensitivity of 36-86%

Raj Cancer 2002 - 222 pts with early biochemical failure after surgery.  174 scans were positive.  False negative rate was 31% in this cohort.  Sensitivity of a subset was 73% and specificity of 53%, PPV 89% - however they utilized response to RT and later progression as the gold standard.

Kahn JCO 1998 - 32pt treated with 60GY postoperatively.  5 negative scans, 18 has bed only, 9 had distant disease.  Prostascint didn't dictate therapy in this case.  Durable clinical response (DCR) was rare in those with distant uptake (27%).  In those with prostate bed uptake DCR had a 62% DCR rate and 88% in those with no uptake.  Some were biopsied in the bed - sensitivity was 78%, specificity 7%.

Proano J Urol 2006 - 44 pts with biochemical recurrence after RP - salvage RT used in all, nodal basins were radiated when Prostascint positive in nodal basins.  41% with a positive scan progressed vs 10% with a negative scan.  Data presented however without KM(Kaplan Meier) estimates.

Nagda IJROBP 2007 - 58 pt with prostascint scan after biochemical recurrence after RP.  Salvage RT used in all.  No difference in any of the prosascint results (distant uptake, bed uptake, or no uptake).  KM estimates used.  Sensitivity 30%, Specificity 58% (using recurrence as the gold standard for presence of disease)

Thomas JCO 2003 - 30 pts with biochemical recurrence after RP.  Prostascint positivity not predicitive of biochemical recurrence.

Koontz IJROBP 2008 - pooled data from some of the above trials - no difference in outcomes dependent on prostascint.

Case outcome - Prostascint results were not utilized in determining treatment - Salvage RT was delivered to 66Gy.  Pt also seen by medical oncology - they agreed with the assessment.  Pt had a complete biochemical response to RT.  Note: patient was treated where scan was negative, and not treated where positive, and a biochemical response was seen.

Monday, February 16, 2009

Brachytherapy Rounds - Sarcoma Implant + Syed Template

Brachytherapy Rounds today covered implants for a sarcoma and a Syed template for vaginal cancer.

Case 1 (High Grade Sarcoma): The sarcoma was a large (24cm in greatest dimension) high grade sarcoma of the R thigh.  The mass was invading skin.  Biopsy confirmed prior to treatment.  He was planned to be treated preoperatively with 50Gy on the IGRT RTOG protocol for sarcomas.  Unfortunately during his external beam treatment had a bleed from his tumor at 32Gy, requiring urgent surgical intervention.

Intraoperatively catheters were placed so that his RT could be completed via HDR. 24 catheters were placed 1cm apart, transversly across the tumor bed.  Catheters extended to the skin on both sides and were tacked down in both the bed and on the skin.  Wound was covered with a wound vac - uncertain to date wether this will heal by secondary intention or by graft at some later date.

Margins on the resection were negative.

HDR dosing was 4Gy x 4 BID prescribed to 1cm in tissue.  Treatment was delayed till 1 week after resection.

Case 2 (Vaginal Adenocarcinoma): Pt with an adenocarcinoma of the RV septum.  Presented with postmenopausal bleeding in 5/2009.  Hx of fibroids, treated with TAHBSO in 2004.  Speculum exam was limited by mass effect in the vaginal and rectum.  EUS was performed revealing tumor extending 10cm along the RV septum extending distally just prior to the introitus.  Biopsy was positive for poorly differentiated adenocarcinoma.

There was a debate at to wether this was a vaginal primary vs. a rectal primary.  Per clinical exam and pathologic review, this was finally characterized as a vaginal primary.

Patient was thought to have a high risk of rectal obstruction during treatment and a diverting colostomy prior to initiation of treatment.  Treated with external beam RT - WPRT with concurrent cisplatin to 50.4Gy.  Pt was initially treated with APPA fields covering the inguinal/pelivic nodes and primary for 10 fx, and was then converted to IMRT to complete treatment.  Pt had a minimal response upon treatment completion.

Pt was then brought in for interstitial implant.  Preplanning was based on pretreatment imaging.  It is important to order the strands in advance and pre-plan placement of the catheters so that one can be ready on the date of implant.

On Date of implant
1. EUA is performed and gold seeds are placed at the proximal, distal and lateral extents of disease.  
2. Cylinder and and template are placed to get a gauge to how far the cylinder can be placed within the vaginal.  This was marked on the cylinder for reference.
3. Clinically it was determined that the catheters should extend 2cm beyond the cylinder.  A sample catheter was placed with the template on the OR table to measure how far from the hub of each catheter should be from the template surface.  (much further than this, one must consider having laparascopic confirmation that the catheters are not entering the peritoneal cavity)
3. 24 catheters were placed posteriorly along the RV septum, and around the cylinder.  After we were happy with the placment the trochars were removed.
4. Patient was brought for CT, and simulator confirmation of placement.
5. Rx was to the 50cGY/hour IDL x 50 hours planned

Unfortunately patient pulled the entire implant out at approximately 36hours.  Uncertain what happened at the time, appartently the patient recieved ativan and became disoriented.  Regardless, the implant was placed in the pig, and the room and patient was surveyed to ensure that all activity was in pig.

Plan is going to be to try to reconstruct the dose delivered and will boost with IMRT afterwards.

Friday, February 13, 2009

EORTC & German trial of Postop RT for Prostate Cancer

In response to a question - here are the two other trials of post op RT for Prostate Cancer referred to in the last post

96-02 (German):
Wiegel ASCO 2005 (abstract only)
385 s/p RP, pT3N0, undetectable PSA
60 Gy post op RT to prostate bed
If PSA detectable, 66Gy postop RT (20%of patients entered)
32 in RT arm did not receive RT (21% of arm)
39 mo follow up
Per protocol: bRFS (4) 81% with RT, 60% NFT (p<0.001)
Intention to treat: bRFS (4) ~75% with RT, 60% NFT (SS)
Grade II rectal toxicity 3%

1,005 Patients underwent RP, node negative with + margins, + ECE, or SVI.
Randomized to
50Gy + 10Gy boost to bed
Close observation (NFT)
Mean f/u 5 years
•90% had PSA <0.2
bPFS (5) 74% RT vs 53% NFT (p<0.001)
Clinical Failure (5) 8.8% RT vs 19.0% NFT (p=0.009)
No difference in OS or Distant Failures
Grade 1-2 toxicity 45%NFT vs 60%RT (SS)
Grade 3 toxicity ~4% for both (NS)

Saturday, February 7, 2009

Adjuvant RT for prostate cancer improves Survival

Dr. Lee draws attention to this article in press at J Urol:

The SWOG trial of 431 men with pT3N0M0 prostate cancer, randomized to 60-64 Gy adjuvant radiotherapy or observation. 70 of the observation group recieved salvage RT at failure.   DM free survival improved with RT (HR 0.71; 95% CI 0.54, 0.94; p = 0.016), and OS improved with RT (HR 0.72; 95% CI 0.55, 0.96; p = 0.023).

This is an extremely important trial, in while adjuvant RT has been demonstrated to improve BRFS in two additional trials (EORTC and a German Trial), this is the first to demonstrate that this translates into a survival benefit.  This is especially relavent when only ~ 15% of patients with pT3 disease are actually referred for RT (unpublished data) nationally.  Duke has done a little better recently with ~35% of patients being referred.