`These records are from CDER’s historical file of information
`previously disclosed under the Freedom of Information Act (FOIA)
`for this drug approval and are being posted as is. They have not
`been previously posted on Drugs@FDA because of the quality
`(e.g., readability) of some of the records. The documents were
`redacted before amendments to FOIA required that the volume of
`redacted information be identified and/or the FOIA exemption be
`cited. These are the best available copies.
`
`

`

`NM! 20451
`
`1 OF 7
`
`

`

`

`

`
`
`

`

`

`

`1
`
`DEPARTMENT OF HEALTH & HUMAN SERVICES
`
`Public Health Service
`
`Food and Drug Administration
`Rockville MD 20857
`
`DEC 27 l995
`
`NDA 20-451
`
`'QLT Phototherapeutics Inc.
`Attention: Mr. Jonathan Kahan
`
`Hogan & Hartson
`555 Thirteenth Street, N.W.
`
`Washington, D.C. 20004-1109
`
`Dear Mr. Kahan:
`
`Please refer to your April 12, 1994 new drug application submitted under
`section 505(b) of the Federal Food. Drug, and Cosmetic Act for Photofrin
`loorfimer sodiun) for Injection, 75 mg vial for use in photodynamic therapy
`with the following devices:
`
`1.
`
`2.
`
`3.
`
`The OPTIGUIDE Fiber Optic Diffuser; and
`
`the Coherent Lambda Plus PDLi and PDLZ Photodynamic Lasers;
`or
`
`the Laserscope Series 600 Dye Modules (Models 630 and 630XP)
`and the Series 700 and 800 KTP/532 or KTP/YAG Surgical Lasers.
`
`We acknowledge receipt of your amendments dated December 4 and 13, 1995.
`
`This new drug application provides for Photofrin for use in photodynamic
`therapy for palliation of patients with completely obstructing esophageal
`cancer, or of patients with partially obstructing esophageal cancer who, in the
`opinion of their physician, cannot be satisfactorily treated with ND—YAG laser
`therapy.
`
`We have completed the review of this application including the submitted draft
`labeling and have concluded that adequate information has been presented to
`demonstrate that the drug product is safe and effective for use as
`recommended in the draft labeling in the submission dated December 13. 1995
`with the revisions listed below. Accordingly, the application is approved
`effective on the date of this letter. We note that the mock versions of the vial
`
`label and carton labeling have not been submitted. This approval does not
`preclude further revision of the final printed labeling, including the vial label and
`cartOn labeling. Please submit these mock versions as soon as possible. As
`discussed by telephone on December 19 and 20, 1995 between Alexandra
`
`

`

`Mancini and Paul Zimmerman of this Division, the revisions to the draft labeling,
`as agreed, are as follows:
`
`Page 2
`
`1.
`
`The drug name, as stated in the titie and first line of the
`DESCRIPTION section and on the first line of the HOW SUPPLIED
`
`section of the package insert, on the vial iabel and on the carton
`labeling, will be PHOTOFRIN' Iporfimer sodium) for Injection.
`
`2.
`
`In the DOSAGE AND ADMINISTRATION section. the sentence,
`
`"However, experience has indicated that mandatory debridement
`may not be necessary due to natural sloughing action in the
`esophagus and may, in fact, neediessly traumatize the area." will
`be changed to, "More recently, experienced investigators have
`indicated that mandatory debridement may not be necessary due
`to natural sloughing action in the esophagus, and may needlessly
`traumatize the area."
`
`3.
`
`In the third line of Pharmacokinetics subsection of the CLINICAL
`
`PHARMACOLOGY section, the word "hour" will be changed to
`"hours".
`
`Additional required changes are indicated on the attached marked-up labeling.
`
`These revisions are terms of the NDA approval. Marketing the product before
`making the revisions, exactly as requested, in the product's final printed
`labeling lFPL) may render the product misbranded and an unapproved new drug.
`
`We remind you of your Phase 4 commitments specified in your submission
`dated December 4, 1995. These commitments, along with any completion
`dates agreed upon, are listed below. Protocols, data, and final reports should
`be submitted to your IND for this product and a copy of the cover letter sent to
`this NDA. Should an IND not be required to meet your Phase 4 commitments,
`please submit protocols, data and final reports to this NDA as correspondence.
`For administrative purposes, all submissions, including labeling supplements,
`relating to these Phase 4 commitments must be clearly designated "Phase 4
`Commitments."
`
`1.
`
`To design and perform a phase 4 singie-arm study to assess the
`efficacy idysphagia response) and safety of PHOTOFRlN-PDT in
`
`patients with partially obstructing esophageal cancer who, in the
`opinion of their pht sician, can not be satisfactorily treated with
`ND-YAG laser therapy. The specifics of the study design and
`patient population will be agreed upon with the Agency prior to
`finalization of the protocol. The limitation of recommended use of
`
`

`

`Page 3
`
`PHOTOFRIN to patients with complete obstructions and partial
`obstructions only where ND-YAG laser cannot be used is based on
`safety concerns. There has not, hovirever, been a trial carried out
`prospectively in the latter population to define effectiveness rates.
`We acknowledge your statement that a draft protocol, which may
`include the collection of pharmacokinetic data in patients as
`described in 2. below, will be provided to the Agency for comment
`in April, 1996;
`
`to conduct phase 4 studies to gather further pharmacokinetic data
`in patients with hepatic impairment and in patients who have
`received more than one course of therapy. Pharmacokinetics will
`also be characterized in male and female patients. We
`acknowledge your statement that this will be completed by April,
`1996:
`
`to develop and validate the capillary electrophoresis (CE) assay,
`capable of fingerprinting the oligomeric mixture of PHOTOFRIN
`(porfimer sodium) for injection, for product release and expiration
`dating. The Agency recognizes the complex nature of this drug
`product and acknowledges that full identification and
`characterization of all the components in the drug product may be
`difficult. However, the responsibility remains for you to
`adequately control and qualify the drug product. We acknowledge
`your statement that this will be completed by October, 1996;
`
`to incorporate a standard in the routine HPLC assay, investigate
`the two wavelengths in the HPLC assay for detection of possible
`nonheme impurities, and validate a test for volatiles (acetic acid).
`We acknowledge your statement that these wiii be completed by
`May, 1996; and
`
`to perform a validation study for the effectiveness of the sterile
`filtration process using PHOTOFRIN. We acknowiedge your
`statement that a final validation report will be available in July,
`1996.
`
`2.
`
`3.
`
`4.
`
`5.
`
`Regarding our suggestions, transmitted in our July 13, 1995 approvable letter,
`to consider retesting the genotoxity of PHOTOFRIN and to consider
`characterizing the mass balance of PHOTOFRIN in humans to the extent
`possible, we acknowledge your statement that such tests will be considered in
`the continuing development plan for PHOTOFRIN.
`
`

`

`Page 4
`
`Please submit fifteen'copies of the FPL as soon as it is available, in no case
`more than 30 days after it is printed. Please individually mount ten of the
`copies on heavy weight paper or similar material. 'For administrative purposes
`this submission should be designated “FINAL PRINTED LABELING" for approved
`NDA 20-451. Approval of this labeling by FDA is not required before it is used.
`
`Should additional information relating to the safety and effectiveness of the
`drug become available, revision of that labeling may be required.
`
`In addition, please submit three copies of the introductory promotional material
`that you propose to use for this product. All proposed materials should be
`submitted in draft or mock-up form, not final print. Please send one copy to the
`Division of Oncology Drug Products and two copies of both the promotional
`material and the package insert directly to:
`
`Food and Drug Administration
`Division of Drug Marketing, Advertising and
`Communications. HFD-240
`5600 Fishers Lane
`
`Rockville, Maryland 2085?
`
`Validation of the regulatory methods has not been completed. At the present
`time, it is the policy of the Center not to withhold apprOVal because the
`methods are being validated. Nevertheless, we expect your continued
`cooperation to resolve any deficiencies that may occur.
`
`Please submit one market package of the drug when it is available.
`
`We remind you that you must comply with the requirements for an approved
`NDA set forth under 21 CFR 314.80 and 314.81.
`
`if you have any questions, please contact:
`
`Paul Zimmerman, Consumer Safety Officer,
`
`(301) 594-5775
`
`Sincerely yours,
`
`M“ {241.655,
`
`Robert Temple, MD.
`Director
`
`Office of Drug Evaluation 1
`Center for Drug Evaluation and Research
`
`

`

`Page 5
`
`cc:
`
`Original NDA 20-451
`HFD-1 SOIDiv. files
`
`HFD-1 50/CSO/PZimmerman
`
`HFD-ZlM.Lumpkin
`HFD-iOO (with labeling)
`HFA-l‘m
`
`HF-2/medwatch (with labeling)
`HFD-BO (with labeling)
`HFD-1 50/GWilliams
`
`HFD-1 50/JRJohnson
`
`HFD-150/AMurgo
`HFD-l SOIYHsieh
`
`HFD-i 50/CHoiberg
`HFD-l 50/RJustice
`
`HFD-150/JDeGeorge
`HFD—1 50/DMcGuinn
`
`HFD—150/RWood
`
`HFD-426/MMehta
`
`HFD-426lARahman
`
`HFD-713/SWilson
`
`HFD-713/CGnecco
`
`HFD-713lAKoutsoukos
`
`HFZ-410lRFeIten
`
`HFD-150/DPease
`
`HFD-160/PCooney
`HFD-‘l 60/CVincent
`
`HFD-643/NSager
`DISTRICT OFFICE
`
`HFD-240/S.Sherman (with draft labeling)
`HFD-638 (with draft labeling)
`drafted: PFZ/December 19, 1995/c:\wpfiies\20451.nda\letters\apletter
`rid Initials: YHsieh/12-20-95
`
`RWood/12-21—95
`
`JDeGeorge for DMcGuinn/12-20-95
`
`JDeGeorge/12-20-95
`AKoutsoukos/12—21~95
`
`AKoutsoukos for CGnecco/12-21-95
`
`ARahman/12-22‘i-95
`
`MMehta/12—21—95
`
`GWilliams/12-20-95
`
`

`

`Page 6
`
`JRJohnson/12-20-95
`
`DPeasa/12- 21- 95
`
`RDeLap/12-22- 95
`
`final type PZimmerman: 12-22-95 Wfé‘x /9/)'°/?r
`APPROVAL W n(151%
`
`

`

`DEPARTMENT OF HEALTH a HUMAN SERVICES
`'5):
`.—...—___._.._._....___—...___...___—_
`
`Putin-.- Health Service
`‘__.___,—_.——_—_._.__——————_p—_—_
`
`Food and Drug Administration
`Rockvilla MD 20867
`
`JUL l 3 5995
`
`NDA 20-451
`
`0LT Phototherapeutics lnc.
`do Mr. Jonathan Kahan
`
`Hogan 81 Hartson
`555 Thirteenth Street, NW.
`
`Washington, D.C. 20004-1109
`
`Attention: Mr. Jonathan Kahan
`
`Dear Mr. Kahan:
`
`Please refer to your April 12, 1994 new drug application submitted under
`section 505th) of the Federal Food, Drug, and Cosmetic Act for Photofrin
`(sterile porfimer sodium) for Injection, 75 mg vial for use in photodynamic
`therapy with the following devices:
`
`1.
`
`2.
`
`3.
`
`The OPTIGUIDE Fiber Optic Diffuser; and
`
`the Coherent Lambda Plus PDL1 and PDL2 Photodynamic Lasers;
`or
`
`the Laserscope Series 600 Dye Modules (Models 630 and 630XP)
`and the Series 700 and 800 KTPI532 or KTP/YAG Surgical Lasers.
`
`The application provides for Photofrin for use in photodynamic therapy for
`palliation of patients with completely obstructing esophageal cancer, or of
`patients with partially obstructing esophageal cancer who, in the opinion of
`their physician, can not be satisfactorily treated with ND-YAG laser therapy.
`
`We also acknowledge receipt of your amendments dated February 3, March 2.
`April 10, May 12, and June 13, 19 and 20, 1995.
`
`We have completed the review of this application as submitted with draft
`labeling. and it is approvable. Before the application may be approved,
`however. it will be necessary for you to submit the following information:
`
`CHEMISTRY
`
`The following comments pertain to your amendment dated March 2, 1995.
`
`

`

`NDA 20-451
`
`Page 2
`
`The current HPLC and mass spectroscopic methods measure little
`more than the total oligomaric content in the mixture. Reference
`10 in your amendment showed size exclusion separation of
`hematoporphyrin diacetate and Photofrin II was achieved on a
`Fractolgel HW-40 (5) column.
`in addition, reference '11 cited in the
`same amendment reported that capillary electrophoresis was able
`to resolve porphyrin oligomers into distinct peaks. Yet, no
`attempts to analyze the drug product oligomeric mixture with
`either method were provided.
`In summary, the current level of
`control and characterization of the drug product is not adequate to
`assure identity, strength, purity and quality, as well as the lot-to-
`lot uniformity of the drug product. We recommend that an
`analytical method that, at a minimum, is capable of fingerprinting
`the oligomeric mixture be developed. This method should be used
`to establish the release specifications of the drug product and
`examine its stability as well.
`(Please refer to Response 7).
`
`The batch records (Table 5) showed that the hematoporphyrin
`dihydrochloride content in the 8 lots range from 78.9% to 88.9%,
`indicating the need for more stringent process control. A
`reasonable approach toward establishing a scientifically sound
`specification for this pivotal intermediate is to determine whether
`lots with higher hematoporphyrin dihydrochloride contents (such
`as PCl414, PC1430, E291191, 3V114BBL and 2182-A-1Pl
`
`produced hematoporphyrin diacetate with more consistent
`compositions.
`If batch records prove this to be the case, we
`
`recommend that the release specifications of hematoporphyrin
`dihydrochloride be established on batch records on those lots.
`
`(Please refer to Response 10).
`
`The responses to Question 13 are not adequate. The justification
`given for the reaction conditions for the synthesis of the drug
`substance, i.e., that the reaction product meets release
`specifications, is not acceptable, because the specifications were
`established on batch history data produced under the very same
`conditions. A suitable justification should include data to show
`that the reaction conditions have been optimized to afford a
`consistent product. For example, in the synthesis of
`hematoporphyrin diacetate, data of a controlled study to show that
`the reaction conditions have been Optimized to maximize the
`
`

`

`NDA 20-451
`
`Page 3
`
`hematoporphyrin diacetate to hematoporphyrin monoacetate ratio
`should be provided.
`
`Please explain why the HPLC protocol specified that the detector
`be set at
`1m lsee page 177, Vol SB) while using peak areas
`measured at 506 nm for calculations.
`In addition, the porfirner
`sodium peak area is calculated as the difference of area counts of
`all detected areas and the sum of peak areas contributed by
`hematoporphyrin, hydroxyethyivinyldeuteroporphyrins and
`protoporphyrin. This practice may tend to exaggerate the actual
`amount of porfimer sodium in the sample. Please explain.
`(Please
`
`refer to response 14).
`
`With regards to the diafiltration process of porfimer sodium,
`justifications to elute the crude material with 120 volume
`replacements of Water for Injection should be provided; it is noted
`that there appears to be a substantial amount of monomeric
`impurities left behind. Furthermore, the flow rate. temperature and
`other conditions such as protection from bright light exposure as
`well as any in process controls other than pH monitoring, should
`be described.
`(Please refer to Response 15L
`
`Specify the proposed additional process controls for the scaled-up
`drying step in the hematoporphyrin diacetate synthesis. {Please
`refer to Response 18).
`
`The responses to Question 21 are not satisfactory. For the
`
`diacetate process. the wide temperature range from 15-30’C
`
`should be justified.
`
`In addition. no information of reaction
`
`completeness tests for both processes were provided.
`
`Due to the difficulty in controlling the oligomerization process and
`characterizing the drug substance mixture, and the absence of a
`validated purification method for the oligomeric mixture, the
`proposed specification of hematoporphyrin diacetate should be
`revised. Batch records of the 12 lots in Table 10 indicated that
`
`as to
`the hematoporphyrin diacetate content ranged from
`.%. Lots PC1012, PC1013, PCl435, PC1436 and PC1437
`
`gave fairly low contents (41.6%, 40.0%, 40.1 %, 41.4% and
`41.1% respectively); lot PC11 12 yielded the most
`hematoporphyrin diacetate at 51.2%; while the remaining 6 lots
`
`
`
`

`

`NDA 20-451
`
`Page 4
`
`(PC1114, PC1115, PC1189, PC1190, PC1194 and PC1422) gave
`more uniform results (47.7%, 45.2%, 43.5%, 46.6%, 46.8% and
`44.4% respectively).
`in theory, one would expect that the
`compositions of porfimer sodium concentrate be directly affected
`by the composition of this monomeric intermediate.
`if the batch
`history records bear out this hypothesis, we recommend that the
`hematOporphyrin diacetata specifications be established on those
`lots that afforded the more consistent drug substance. Further
`analysis of the production history records may help identify
`
`parameters in the manufacture of porfimer sodium bulk
`concentrate that most affect the composition of the product.
`Finally, batch records should be submitted to justify the proposed
`0.8% limit for biomine content. (Please refer to Response 21).
`
`9.
`
`Considering the current relatively wide release specifications of the
`drug substance and the drug product. the practice of designating a
`production run as the reference standard without any purification
`and replacing it with a new iot when the first one expires results in
`the use of a reference material with its purity profile changing
`periodically for routine comparison of production batches. We
`recommend that specifications be established for the drug
`substance reference standard.
`(Please refer to Response 23).
`
`10.
`
`Please identify the peaks eluted at 1.517 min, 5.733 min, 6.917
`min, 13.183 min and those eluted between 19.8 min and 20.183
`
`min in the HPLC chromatogram of internal standard 5745 (Figure
`3, page 50} and explain how to differentiate these peaks from
`those from Photofrin eluted at the same time. The major
`component in Photofrin standard 8730 (Figure 2, page 491
`exhibited a retention time at 19.91? min, whereas the major
`Photofrin peak in the HPLC trace provided previously (page 1 14,
`Vol 1.3) gave a retention time around 15.“) min. Please identify
`whether this significant shift in retention time is a result of change
`in column conditions or in mobile phase.
`(Please refer to Response
`271.
`
`1 1. We have the following comments for Response 28:
`
`a.
`
`The question regarding the possible formation of olefinic
`degradation products during the manufacturing process and
`upon storage is not addressed.
`
`

`

`NDA 20-45]
`
`Page 5
`
`b.
`
`c.
`
`Please provide data to show that the UWVlS method
`(absorption at
`nm) to assure the photostability of the
`drug product has been validated.
`
`An additional release specification has been proposed for
`Photofrin to check its relative absorption bands at
`and
`nm. Specify the proposed limits of the release
`specification and test procedures.
`
`d.
`
`Data should be submitted to demonstrate that the current
`
`HPLC method is capable of detecting degraded products of
`3% H202-stressed Photofrin.
`
`e.
`
`nm)
`The UV/VlS spectroscopic method
`establishing Photofrin stability towards oxidation, and the
`limit of detection should be submitted.
`
`12.
`
`Because of the method of storage used for the bulk porfimer
`sodium, freeze~thaw studies should be conducted as soon as
`
`possible. Please refer to Deficiencies 1 and 1 1 cited in this letter
`
`for comments on UV/VIS methods, the mass spectroscopic
`method and HPLC assay.
`
`The stability of the drug product when stressed with heat.
`peroxides and bright light should be properly examined after
`suitable stability-indicating analytical methods have been
`developed.
`(Please refer to Response 30).
`
`13.
`
`The proposed ranges for the trimer and tetramer ion ratio levels
`relative to dimers are acceptable. No response has been made to
`
`the Agency's request to develop more accurate methods for
`determining the oligomer size distribution.
`(Please refer to
`Response 39h).
`
`14.
`
`Sections of the NDA allude to olefinic trimers and tetramers being
`responsible for some of the more significant adverse events.
`Response 39i is not acceptable. Analytical methods to preperly
`characterize and assay olefinic oligomers in the drug product
`mixture should be developed.
`
`15.
`
`Please provide data. including monomer impurities, oligomer
`
`

`

`NDA 20-451
`
`Page 6
`
`content and composition, ester/ether linkage ratios, olefin side
`chain content and hydroxyethyl side chain content to demonstrate
`that the preclinical batches, clinical batches and those produced
`for commercial distribution are equivalent in chemical composition
`and biological activity. The release Specifications and shelf-life
`specifications, as described in Monographs 18351 and 19857,
`should be revised after a suitable analytical method. one that can
`fingerprint the oligomeric mixture, is developed.
`In addition,
`because of the adverse events related to olefinic trimers and
`
`tetramers, the Photofrin release specification should include an
`olefin content limit as well.
`(Please refer to Response 44).
`
`16.
`
`We recommend testing intervals of 3, 6, 9, 12, 18, 24 and 36
`months for long term stability study and that the assay method
`employed in the protocol should be validated as stability-indicating
`less Comment 1). Cumulative stability data on production batches
`
`should be submitted to FDA as part of the annual report required
`under 21 CFR 314.81ibli2).
`(Please refer to Response 46)
`
`The following comments pertain to your amendment of May 12, 1995.
`
`The HPLC assay as described in this submission is not adequate to replace the
`mouse bioassay currently used to assure the efficacy of PHOTOFRIN batches.
`
`Chemistry comments:
`
`1.
`
`to
`
`Please provide justifications for selecting the 3 types of samples
`that were used in the development of the bioactivity-indicating
`HPLC assay (paragraph 2 , page 11).
`
`This HPLC assay failed to separate monomeric irr-nurities from the
`
`in addition, it was reported that area
`solvent front (Region 1).
`increases in Region 3 were observed in HPLC profiles for all
`stressed Photofrin dry powder samples, indicating that this method
`is not able to resolve the degraded products from their parent
`compounds. Justifications should be provided for selecting the
`described HPLC conditions for release testing.
`
`Development of the bioactivity-indicating HPLC assay was only
`supported by bioassay date and HPLC chromatograms. The
`bioassay is a pass/fail functional test and HPLC profiles are poorly
`
`

`

`NDA 20-451
`
`Page 7
`
`resolved. To deveiop an alternative to the bioassay, other
`analytical data. including UV/VIS and‘mass spectroscopic data,
`should be provided as a measure of the chemical integrity of the
`oligomeric mixture.
`
`Justification should be provided for the proposed acceptable range
`of area percent values of peaks in Regions 1 and 3 for bioactive
`Photofrin sampies.
`
`Table 4 (page 16) gave the bioassay results of reconstituted
`samples stored frozen at -20"C.
`It should be noted that the
`stabiiity of the reconstituted samples under the prescribed condi-
`tions has not been fully established. The length of time that the
`reconstituted samples had been stored should be specified and
`other data to support the chemical identity of the oligomeric
`mixture submitted.
`
`In one of the stress studies of lyophilized Photofrin, dose response
`curves indicated a reducticn of EDso from
`mg/kg to
`mg/kg
`for the Photofrin vials that had been :reated at 80°C in the dark or
`
`40°C under white light (NDA 20-451, Volume SB, Section
`3.2.5.1.2, page 197). However, in a second experiment, when
`the samples were similarly stressed, no bioactivity enhancement of
`the treated samples was detected (Study Report AM—93104S,
`
`Appendix 2, page 87). Piease explain.
`
`Pharmacology comments:
`
`7.
`
`Why are two peaks in the PHOTOFRIN chromatogram on page 10
`and elsewhere marked Hi"1roxyethylvinyldeuteroporphyrin (HVD)?
`Are two simiiar peaks eluted with this system when HVD
`standards are chromatographed?
`
`The scale of the chromatograms in the submission varies. Was
`
`the PHOTOFRIN sample concentration standardized for the HPLC
`separations?
`
`Were the batches used to define and validate the BIHPLC assay
`manufactured under scheme l,
`IR or ll?
`
`10.
`
`You could have prepared the Region 3 reduced sample (appendix
`
`

`

`NDA 20-45]
`
`Page 8
`
`11.
`
`12.
`
`2, page 97) by HPLC purification of PHOTOFRIN similar to the
`preparation of the Region 3 enrichedsample. Hp and HVD
`together do not appear to account for most of the area in R2. The
`contribution of the other peaks in PHOTOFRIN chromatographic
`regions R1 and R2 cannot be determined. The anaiysis assumes
`there is no efficacy threshold for some other component of R1 or
`R2 (non-linear dose response) and that Hp and HVD are completely
`ineffective. The latter of these two assumptions is not well
`,
`supported, since this sample had an unexpectedly low E05,,
`mg/kg).
`If no PHOTOFRIN activity is associated with Region 2,
`the intercept of the linear analysis should be statistically equal to
`0. Yet, this intercept value is
`(close to the ”ED“ value of the
`Region 3 reduced sample).
`
`The two points from the Region 3 manipulated samples have an
`inappropriately large influence on your linear analysis. They are far
`outside the range of the other points and are tin-weighted. Also
`the point at E050
`%area =
`in the photolysis
`experiment appears an outlier only when the data are analyzed
`using the reciprocai of E050. When E050 is plotted directly against
`°/oarea. this point if; an important part of the dose response.
`
`Please explain why this point was excluded.
`
`If the change in ‘llEDso from these PHOTOFRIN samples results
`from the variation of a single linear parameter, the slopes derived
`from the three experiments should be statistically indistinguishable.
`The individual slopes are
`‘ photolysis},
`(pyrolysis) and
`lR3-manipulated). The Student's-t statistic for the slopes
`from the photolysis and pyrolysis data is significantly larger than
`the tabular t value indicating that the slopes are significantly
`different. These lines are not parallel. The slopes and intercepts
`for all three lines appear from the graph to be different. Please
`explain why these experiments were combined for this analysis?
`A linear analysis can sometimes be used to describe a dose
`
`response within the pseudo~linear region of the curve when the
`curve is well defined. Nevertheless, the photolysis experiments do
`not vary %area R3 across a sufficiently wide range to adequately
`define the upper and lower limits of a dose response curve. The
`pyrolysis experiments demonstrate no dose response and the
`manipulated R3 samples are probably not measuring the same
`pseudo-linear phenomenon. The relationship between EDso and
`
`

`

`NDA 20-451
`
`Page 9
`
`13.
`
`14.
`
`15.
`
`%area R3 wiii probably be better analyzed as a reverse-sigmoidal
`dose response where integrated %area is the dose and ED50 is the
`response. A replot of the raw data from the pyrolysis experiment
`suggests such a relationship. You should use such a dose
`response analysis to specify acceptable limits for the %area from
`lower (E050 = x mglkg) and upper (E050 = y mg/kgl dose
`response values.
`
`Since a dose response relationship probably exists between some
`component of Region 3, what is the rationale for using three
`standard deviations from the batch-to-batch mean %area Region 3
`to define the limits for batch acceptability? How do these limits
`relate to the dose response? Batches P91-164 and P91-163 had
`%area Region 3 values of 32.3 and 31.1 respectively in the
`experiments used to calculatethis batch-to-batch mean (Appendix
`1, page 81). These batches had %area Region 3
`'ues of 35 and
`36 respectively in subsequent experiments to determine the dose
`response (Appendix 2. page 102}. What caused this difference?
`Using the limits defined from the batch-to-batch mean and
`
`standard deviation of Appendix 1, samples with only 2% higher
`%area Region 3 values than these obtained for P91-164 and P91-
`163 in Appendix 2 would be reiected.
`
`In future submissions please include the raw data from the E050
`experiments. How many points did you use to determine E050?
`How many animals were in each dose group?
`
`The results of these PHOTOFRIN degradation studies (photoiysis
`and pyrolysis) imply that Region 3 of the chromatogram contains
`at least two components that significantly affect the integrated
`%area. The concentration of one component increases with
`pyrolysis of the lyophilized powder yet does not influence PDT
`efficacy. That of another decreases with photolysis of
`reconstituted PHOTOFRIN and is central to the mechanism of PDT.
`
`The current HPLC assay cannot distinguish between these two
`components and thus cannot predict batch-to-batch efficacy or
`uniformity. For example, a batch exposed to excess heat might be
`rejected because the %area of Region 3 was to great, yet the
`batch would pass the bioassay This finding is not discouraging.
`suggests that better chromatographic separation or defining
`smaller chromatographic regions may distinguish the component
`
`it
`
`

`

`NDA 20-451
`
`Page 10
`
`within R3 that correlates with activity. This method cannot as yet
`replace the bioassay used to assure the efficacy of each batch of
`PHOTOFRIN, but such a replacement is probably attainable.
`
`MICROBIOLOGY
`
`Please provide the additional or clarifying information, described below. to
`support the sterilization process validation information portion of your NBA and
`fulfill the post-approval commitments. Although you have cited the FDA's
`"GUIDELINE FOR SUBMITTING DOCUMENTATION FOR STERILIZATION
`
`PROCESS VALIDATION IN APPLICATIONS FOR HUMAN AND VETERINARY
`
`DRUG PRODUCTS (DECEMBER 3. 1993)" in 4, 05-1 3-94 amendment to the
`NDAI we request that you use this cited format to prepare your response to
`these requests for further information. Manufacturing operations, filling
`process, monitoring activities, batch records, physical lay-out (and other items
`below) need to be more completely described in order to complete the review
`and evaluation of the microbiological quality and sterility assurance of the
`subject drug product.
`
`if these types of information have been provided to more recently prepared
`NDAs. the information may be referenced. Please provide for a 'desk copy' for
`microbiology review purposes.
`
`The following concern your descriptions for ProductWas;
`
`a.
`
`b.
`
`What is the pH of the diluted sample in NaOH, and
`
`is the pH of the resulting sample solution adjusted to
`neutraiity prior to filtration for the microbial limits test?
`
`2.
`
`Part of the manufacturing procedure for profimer sodium bulk
`concentrate {Vol 3, p 195496) includes filtering through a
`
`p
`
`

`

`NDA 20-45}
`
`Page 1 1
`
`
`
`filter into a polycarbonate bottle. The resulting product solution is
`then tested by
`_
`(Vol 4, p 1-1 5) whose
`specification for 'Microbial Limit' is not more than
`CFU/ mL of
`concentrate. Please define exactly what the 'sampie“ is that is
`being tested for the microbial limit by both Methods
`of
`
`Vol 4, p 1-15)
`,
`_
`The reference cited for
`namely, a 'Special Microbiological Study Report: PHOTOFRIN 1 1
`Bulk Liquid, Qualification of Total Aerobic Count: dated 1 .4.94
`[issued to qualify the cited method identified as "Microbial Limits"
`(Vol 4, p 4)] should be provided to clarify the microbiological
`methods used for Photofrin lporfimer sodium).
`
`The following concern yourWWW
`
`Some additional information is necessary. Also, the information provided in
`Generai Emma (Attachment 3, Section 3.2.3.5) does not clearly
`describe the manufacturing and filling process or monitoring activities
`sufficiently to completely evaluate the microbiological quality and sterility
`assurance of the subject drug product.
`
`1.
`
`The specific acceptance criteria for the validation studies
`conducted for the
`
`should be
`_
`provided. Summaries of the sterilization process validation
`protocols, studies, and data that specify the acceptance criteria,
`and demonstrate that the acceptance criteria were met, should be
`provided to the application.
`
`All product-contact equipment needs to be clearly identified by
`name and room location. The product, personnel, and component
`flow sequence from bulk formulation to filiing to packaging needs
`to be provided.
`
`The following concern the 5.5mm and
`Environmental monitoring program.
`
`1.
`
`Please provide information regarding the specific room and
`locations within the room where samples are obtained. A sketch
`or diagram (blueprints are not necessary) with the sampling
`
`

`

`NDA 20-451
`
`Page 12
`
`location indicated would be helpful.
`
`The equipment used in the aseptic fill manufacturing process
`needs to be adequately identified and described in the application
`so that samples obtained from (or near?l product-contact
`equipment" are clearly understood.
`
`Microbiological protocols, methods, and data generated by this
`type of environmental monitoring should be provided. For
`example, quantitative limits are given for swab samples obtained
`from surfaces. but there is no information to indicate that the
`
`'swab sample' is either obtained or cultured in a quantitative
`fashion.
`
`There appear to be occasional surveys for ‘mold‘ and 'anaerobes‘
`(p 14, 15) but no